WO2019102277A1

WO2019102277A1 - Method and system for determining hematological parameters in a peripheral blood smear

Info

Publication number: WO2019102277A1
Application number: PCT/IB2018/056315
Authority: WO
Inventors: Dheeraj MUNDHRA; Bharath Cheluvaraju; Tathagato Rai Dastidar; Apurv Anand; Rohit Kumar Pandey; Himanshu Sharma
Original assignee: Sigtuple Technologies Private Limited
Priority date: 2017-11-23
Filing date: 2018-08-21
Publication date: 2019-05-31

Abstract

The present disclosure relates to a method and system for determining hematological parameters in a Peripheral Blood Smear (PBS). The blood analyzer receives a plurality of images from the monolayer of the PBS. Each of the plurality of images are processed by the system for extracting a plurality of image patches from each of the plurality of images. The system identifies RBCs in each of the plurality of image patches using deep learning models. The system determines a plurality of features required for calculating volume of the RBC in each of the plurality of image patches. The system classifies each RBC into predefined types based on area of the RBC, area of pallor region and area of the non-pallor region. Finally, the system identifies a number of RBCs belonging to each of the predefined type and determines the hematological parameters based on the number of RBCs of each pre-defined type.

Description

“METHOD AND SYSTEM FOR DETERMINING HEMATOLOGICAL PARAMETERS IN A PERIPHERAL BLOOD SMEAR”

FIELD OF THE DISCLOSURE

The present subject matter relates to study of blood cells/ hematology. The present subject matter relates more particularly, but not exclusively, to an image processing method and an image processing system for determining hematological parameters in a peripheral blood smear.

BACKGROUND

Complete blood count (CBC) is a basic screening hematology test. A CBC report consists of hematological parameters like hemoglobin content in a sample, count of blood cells per unit volume, Mean Corpuscular Volume (MCV), Mean Corpuscular Hemoglobin (MCH) etc. Abnormally high or low variations in the hematological parameters may indicate the presence of many forms of diseases. Hence, hematological parameter estimation is among the most commonly performed blood tests in the field of hematology, as they can provide an overview of general health status of a patient. Few existing methodologies use hematology analyzers for determining various hematological parameters from a blood sample. Hematology analyzers use techniques such as flow cytometry to estimate hematological parameters etc., in a blood sample. However, the hematology analyzers measure the type of blood cell by analyzing data about the size and aspects of light as they pass through the cells. The amount of light scattered by the blood cells are mainly dependent on the size of the blood cells. Hence, certain abnormal cells in the blood sample may not be identified correctly, requiring manual review of the instrument's results for identification of any abnormal cells which could affect the estimation of hematological parameters. Moreover, the analyzers mentioned above use reagents in every analysis, thereby, increasing the cost.

Few other existing methodologies involve use of Hemocytometers (counting chambers that hold a specified volume of diluted blood and divide it with grid lines) for calculating the volume of blood cells and other hematological parameters. This technique depends on manual counting and manual counting is subjected to sampling error because only few cells are counted compared to automated analysis.

Some other existing methodologies illustrate the estimation of volume and hemoglobin content of blood cells using image-based analysis, by capturing images of the blood smear. Further, the existing methodologies use estimators for volume of each cell. A mean of the estimators across all the cells are computed for estimating the hematological parameters. The existing methodologies are applicable when the stain in blood smear is constant and the blood smear as well as the stain quality is maintained using high cost devices, hence leading to an overall high cost of the system and reagents. Further, statistical parameters such as mean of estimators for all the cells may not necessarily represent the hemoglobin content of all the cells in the entire blood sample, because the Red Blood Cells (RBCs) are present in aplenty and may have varied size and shapes m in a standard blood smear. The methodology is specifically applicable in cases where the estimators of hemoglobin content in cells of a blood sample does not follow a gaussian distribution. Further, the existing methodologies estimate hematological parameters by capturing random Field of Views (FoVs) of the blood smear. Estimation of hematological parameters using random FoV results in a biased estimation of hematological parameters.

The information disclosed in this background of the disclosure section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.

In one embodiment, the present disclosure discloses a method for determining hematological parameters in a blood smear. The method comprises receiving a plurality of images of a blood smear from an image capturing unit configured to focus and capture images of the blood smear. The plurality of images is captured from a monolayer of the blood smear. Each image of the plurality of images wherein each image is processed for extracting a plurality of patches. Each patch comprises a Red Blood Cell (RBC). Further, the method comprises determining a plurality of features required for calculating volume of the RBC in each patch. A value of a first feature from the plurality of features is determined by calculating a difference between an area of the RBC and a normalized value. The area of the RBC comprises an area of a pallor region in the RBC and an area of a non pallor region m the RBC and the normalized value indicates the area of pallor region. Computing the normalized value for each patch comprises determining a value of intensity of each of a plurality of pixels in each patch. A first set of pixels among the plurality of pixels indicates a peak value of intensity in a background region and is characterized as background peak and a second set of pixels among the plurality of pixels indicate a peak value of intensity in a foreground region and is characterized as foreground peak. Further, computing the normalized value comprises determining a difference between the value of intensity of each pixel in the patch and an intensity threshold value determined based on the foreground peak and the background peak. Thereafter, determining a distance between the foreground peak and the background peak in the patch and determining the normalized value based on a ratio of the determined difference and the distance. The hematological parameters are determined based on the plurality of features.

In an embodiment, the present disclosure relates to a blood analyzer for determining hematological parameters in a blood smear. The blood analyzer comprises a processor and a memory. The memory is communicatively coupled with the processor and stores processor executable instructions. The processor is configured to receive a plurality of images of a blood smear from an image capturing unit configured to focus and capture images of the blood smear. The plurality of images is captured from a monolayer of the blood smear. Each image of the plurality of images wherein each image is processed for extracting a plurality of patches. Each patch comprises a Red Blood Cell (RBC). Further, processor is configured to determine a plurality of features required for calculating volume of the RBC in each patch. A value of a first feature from the plurality of features is determined by calculating a difference between an area of the RBC and a normalized value. The area of the RBC comprises an area of a pallor region in the RBC and an area of a non pallor region in the RBC and the normalized value indicates the area of pallor region. Computing the normalized value for each patch comprises determining a value of intensity of each of a plurality of pixels in each patch. A first set of pixels among the plurality of pixels indicates a peak value of intensity m a background region and is characterized as background peak and a second set of pixels among the plurality of pixels indicate a peak value of intensity in a foreground region and is characterized as foreground peak. Further, the normalized valise is computed by determining a difference between the value of intensity of each pixel in the patch and an intensity threshold value determined based on the foreground peak and the background peak. Thereafter, determining a distance between the foreground peak and the background peak in the patch and determining the normalized value based on a ratio of the determined difference and the distance. The hematological parameters are determined based on the plurality of features.

The foregoing summary is illustrative only and is not intended to be in any way limiting. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features will become apparent by reference to the drawings and the following detailed description.

BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS

The novel features and characteristic of the disclosure are set forth in the appended claims. The disclosure itself, however, as well as a preferred mode of use, further objectives and advantages thereof, will best be understood by reference to the following detailed description of an illustrative embodiment when read in conjunction with the accompanying figures. One or more embodiments are now described, by way of example only, with reference to the accompanying figures wherein like reference numerals represent like elements and in which:

Figure 1 shows a blood analyzer for determining hematological parameters in a Peripheral Blood Smear (PBS) in accordance with some embodiments of the present disclosure;

Figure 2 shows an exemplary block diagram of a blood analyzer for determining hematological parameters in a Peripheral Blood Smear (PBS) in accordance with some embodiments of the present disclosure; Figure 3 shows an exemplary flowchart illustrating method steps for determining hematological parameters m a Peripheral Blood Smear (PBS)m accordance with some embodiments of the present disclosure;

Figure 4 shows an exemplary flowchart illustrating method steps for estimation of the value of a first feature indicative of volume after mean centering and contrast normalization, in accordance with some embodiments of the present disclosure.

Figures 5A, SB, SC, 5D and show exemplary images of a PBS in accordance with some embodiments of the present disclosure;

Figures 6.4 and 6B show exemplary images of intensity graphs generated for different Field of Views (FoVs) of a PBS used for estimating a value of a plurality of features, in accordance with some embodiments of the present disclosure;

Figure 8A and Figure 9A show exemplary images of different Field of Views (FoVs) of a PBS indicative of modifications in the FoV s due to computation of the plurality of features, in accordance with some embodiments of the present disclosure; and f i re 7 A, f igure /if, f igure 811, f igure SL, igure PIS, Figure 9 , f igure 10 A and Figure 10B show intensity v/s frequency histograms generated based on the plurality of features, in accordance with some embodiments of the present disclosure;

Figure I I illustrates a block diagram of an exemplary computer system for implementing embodiments consistent with the present disclosure.

It should be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative systems embodying the principles of the present subject matter. Similarly, it will be appreciated that any flow charts, flow diagrams, state transition diagrams, pseudo code, and the like represent various processes which may be substantially represented in computer readable medium and executed by a computer or processor, whether or not such computer or processor is explicitly shown. DETAILED DESCRIPTION

In the present document, the word "exemplary” is used herein to mean "serving as an example, instance, or illustration. " Any embodiment or implementation of the present subject matter described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

While the disclosure is susceptible to various modifications and alternative forms, specific embodiment thereof has been shown by way of example in the drawings and will be described in detail below. It should be understood, however that it is not intended to limit the disclosure to the particular forms disclosed, but on the contrary, the disclosure is to cover all modifications, equivalents, and alternative falling within the scope of the disclosure.

The terms“comprises”,“comprising”, or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a setup, device or method that comprises a list of components or steps does not include only those components or steps but may include other components or steps not expressly listed or inherent to such setup or device or method. In other words, one or more elements in a system or apparatus preceded by “comprises... a” does not, without more constraints, preclude the existence of other elements or additional elements in the system or apparatus.

In the following detailed description of the embodiments of the disclosure, reference is made to the accompanying drawings that form a part hereof, and in which are shown by way of illustration specific embodiments in which the disclosure may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the disclosure, and it is to be understood that other embodiments may be utilized and that changes may be made without departing from the scope of the present disclosure. The following description is, therefore, not to be taken in a limiting sense.

Embodiments of the present disclosure relate to a method and system for determining hematological parameters in a Peripheral Blood Smear (PBS). The blood analyzer receives a plurality of images from the monolayer of the PBS. Each of the plurality of images are processed by the system for extracting a plurality of image patches from each of the plurality of images. Further, the system identifies RBCs in each of the plurality of image patches using deep learning models. Thereafter, the system determines a plurality of features required for calculating volume of the RBC in each of the plurality of image patches. Furthermore, the system classifies each RBC into predefined types based on area of the RBC, area of pallor region and area of the non-pallor region. Finally, the system identifies a number of RBCs belonging to each of the predefined type and determines the hematological parameters based on the number of RBCs of each pre-defined type.

Figure 1 shows a blood analyzer 100 for determining hematological parameters in a Peripheral Blood Smear (PBS) in accordance with some embodiments of the present disclosure. The blood analyzer 100 comprises a microscopic system 101, Peripheral Blood Smear (PBS) 102, and a user interface 103. The PBS is a thm layer of blood corresponding to a blood sample, smeared on a glass microscope slide and then stained m such a way as to allow^r the various blood cells to be examined microscopically. The PBS 102 may correspond to a blood sample of a subject. In an embodiment, the subject may be a patient or any living being. An image capturing unit of the microscopic system 101 may capture high resolution images or enhanced microscopic images of the PBS 102. A plurality of images of the PBS 102 may be captured by the imaging unit of the microscopic system

101. The plurality of images may be referred as PBS images hereafter in the present disclosure. In an embodiment, the PBS images may be of a monolayer region of the PBS

102. The blood analyzer 100 receives the PBS images and processes the PBS images in order to determine the hematological parameters in the PBS. Each of the PBS images received from the microscopic system 101 may be a RGB color image. Each of the PBS images are processed by the blood analyzer 100 for extracting a plurality of image patches from each of the PBS images. The blood analyzer 100 may convert each of the PBS images into a greyscale image during the processing. Further, the system identifies RBCs in each of the plurality of image patches using deep learning models. Thereafter, the blood analyzer 100 determines a plurali ty of features required for calculating volume of the RBC in each of the plurality of image patches. The blood analyzer 100 classifies each RBC into predefined types. The predefined types may be based on area of the RBC, area of pallor region and area of the non-pallor region. The blood analyzer 100 identifies a number of Red Blood Cells (RBCs) belonging to each of the predefined type and determines the hematological parameters based on the number of RBCs of each pre-defined type. The determined hematological parameters are provided to the user interface 104, winch may provide an indication of the determined hematological parameters to a clinician or any person analyzing the PBS images. in an embodiment, the PBS images, may include, the plurality of images of the PBS 102 m the monolayer of the PBS 102. The PBS images may be retrieved from the microscopic system 101. The microscopic system 101 may be any system winch is configured to retrieve microscopic images of the PBS 102 and provide the microscopic images of the PBS to the blood analyzer 100. In an embodiment, the microscopic system 101 may comprise a microscope, a stage and the image capturing unit for retrieving enhanced microscopic images of the PBS 102. The stage may be configured to hold the PBS 102. The microscopic device 101 may be configured to focus on region of interest in the PBS 102. The PBS has three main regions namely, clumped region, the monolayer region, and feather edge region. Density of blood decreases in the order of clumped region, monolayer region and feather edge region. Therefore, the PBS images acquired from the clumped region and the feather edge region of the PBS results m over or under estimation of hematological parameters respectively. In the monolayer region, all the cells are well separated or slightly touching each other, which result in unbiased estimation of hematological parameters. Therefore, the microscopic system 101 may optimally scan the PBS to determine the monolayer region of the PBS 102. The image capturing unit may be configured to capture enhanced microscopic images of the PBS 102 in the monolayer region.

In an embodiment, the PBS images may comprise plurality of images, for example 120 images covering the monolayer region of the PBS 102. The formats of the type of PBS images may be one of, but not limited to. Resource Interchange File Format (RIFF), Joint Photographic Experts Group (JPEG/JPG), BitMaP (BMP), Portable Network Graphics (PNG), Tagged Image File Format (TIFF), Raw image files (RAW), Digital Imaging and Communication (DICOM), Moving Picture experts group (MPEG), MPEG-4 Part 14 (MP4), etc.

In an embodiment, the user interface 103 may comprise a display device, a report generation device or any other device capable of providing a notification or interacting with a user. The user interface 103 may be used to notify the determined hematological parameters to a clinical specialist examining the PBS images. In an embodiment, the user interface 103 may be a part of the blood analyzer 100 or may be associated with the blood analyzer 100.

In an embodiment, the display device may be used to display the hematological parameters determined by the blood analyzer 100. The display device may be one of, but not limited to, a monitor, a Liquid Crystal Display (LCD), a Light Emitting Diode (LED) display and/or any other module present which is capable of displaying the hematological parameters.

In an embodiment, the report generation unit may be used to generate a report comprising details of the hematological parameters determined by the blood analyzer 100.

In an embodiment, the microscopic system 101 may be separate unit associated with the blood analyzer 100. The PBS images captured by the microscopic system 101 may be provided as an input to the blood analyzer 100 via a communication interface, for example wired and wireless communication interfaces.

Figure 2 shows an exemplary block diagram of a blood analyzer 100 for estimating hematological parameters in the PBS 102, in accordance with some embodiments of the present disclosure. The blood analyzer 100 may include at least one processor 203 and a memory 202 storing instructions executable by the at least one processor 203. The processor 203 may comprise at least one data processor for executing program components for executing user or system-generated requests. The memory 202 is communicatively coupled to the processor 203. The blood analyzer 100 further comprises an Input/ Output (I/O) interface 201 The I/O interface 201 is coupled with the processor 203 through which an input signal or/and an output signal is communicated. In an embodiment, the I/O interface 201 may provide the PBS images to the blood analyzer 100. In another embodiment, the I/O interface 201 couples the user interface 104 to the blood analyzer 100.

In an embodiment, the processor 203 may implement machine learning models for analyzing the PBS images. The processor 203 may implement any existing machine learning models which may include, but are not limited to, decision tree learning, association rule learning, artificial neural networks or deep learning, inductive logic programming, support vector machines, clustering, Bayesian networks, reinforcement learning, representation learning, similarity and metric learning, genetic algorithms, rule- based machine learning.

In an embodiment, data 204 may be stored within the memory 202. The data 204 may include, for example, training data 205, image data 206, features data 207 and other data 208.

In an embodiment, the machine learning models may be trained to analyse the PBS images using the training data 205. The training data 205 may comprise few PBS images from the PBS images. In an embodiment, for training the blood analyzer 100, random patches are extracted from each of the few' PBS images. Further, a blood cell in each of the random patches are labeled as one of RBC and non-RBC by experts based on various parameters related to the blood cell. The labeled patches are used for training the deep learning model. For instance, patch 1 is extracted from a PBS image. The blood cell in patch 1 is found to be a RBC. Hence, the patch 1 is labeled as RBC by the expert. The patch 1 is used for training the blood analyzer 100. Further, when the blood analyzer 100, encounters a patch 2 similar to the patch 1, it may automatically classify the blood cell in patch 2 as RBC. The blood analyzer 100 may be trained using a vast set of images from the training data 205. Thereby, the blood analyzer 100 may be able to efficiently identify RBCs in the plurality of image patches extracted from each of the PBS images. The blood analyzer 100 may discard the White Blood Cell (WBC) and platelets and any overlapping cell in each of the PBS images. In an embodiment, the image data 206 refers to the properties of each of the PBS images. The properties may include, but are not limited to, resolution or quality of the PBS images, sharpness of the PBS images, image size, and image format. In an embodiment, the image data may also comprise the plurality of image patches extracted by the blood analyzer 100.

In an embodiment, the features data 207 refers to the plurality of features determined by the blood analyzer 100 for each of the plurality of patches comprising the RBC in each of the PBS images. The features data 207 stores value corresponding to each feature of the plurality of features computed for each of the PBS images.

In an embodiment, the other data 208 may include, but is not limited to weighing parameters data and clusters data. The -weighing parameters data refers to different parameters for assigning weight to each of the PBS images. The weighing parameters data may be based on data present in the image data 206. Each of the PBS images may be assigned a weight based on one or more parameters present in the weighing parameters data. The cluster data may include the predefined types of RBCs considered. The predefined types may be determined based on one of on area of the RBC, area of pallor region and area of the non-pallor region. For instance, one of the predefined type may be a RBC with low area value and high value of pallor ratio (ratio of area of pallor region and area of non-pallor region). Based on the value of the plurality of features determined for each of the RBCs, a RBC may be classified as belonging to one of the pre-defined types.

In an embodiment, the data 204 in the memory 202 is processed by modules 209 of the blood analyzer 100. As used herein, the term module may refer to an application specific integrated circuit (ASIC), an electronic circuit, a field-programmable gate arrays (FPGA), Programmable System-on-Chip (PSoC), a combinational logic circuit, and/or other suitable components that provide the described functionality. The modules 209 when configured with the functionality defined in the present disclosure will result in a novel hardware. In one implementation, the modules 209 may include, for example, an image acquisition model 210, an image patch extraction module 211, a feature determination module 212, a hematological parameter determination module 213 and other modules 214. It will be appreciated that such aforementioned modules 209 may be represented as a single module or a combination of different modules.

In an embodiment, the image acquisition model 210 acquires the PBS images (enhanced images of the PBS) from the microscopic system 101 , for processing and for estimating hematological parameters. The image acquisition model 210 may convert each of the PBS images into one of a binary image or a grayscale image for further processing. The resulting images are one of a binary form and grayscale form of each of the PBS images. The binary form or greyscale form of each of the PBS images may be referred to as plurality of processed images hereafter in the present disclosure.

In an embodiment, the image patch extraction module 211 may extract and segment plurality of image patches from each of the plurality of processed images. Each of the plurality of image patches comprises the blood cell. The blood cell may be one of RBC, WBC and platelet. The area of the plurality of image patches may be pre-defmed. In an embodiment area of the blood cells in the PBS may be considered as a patch and may be extracted. In an embodiment, the image patch extraction module 21 1 extracts plurality of image patches, such that each of the plurality of image patches comprises a blood cell of size of RBC.

In an embodiment, the image patch extraction module 21 1 may classify the blood cell in each of the plurality of image patches into one of RBC and non-RBC. The extraction of the plurality of image patches may be performed using pre-trained Artificial Intelligence (AI) models or deep learning models. The pre-trained AI models may be a combination of convolutional neural networks and statistical models. Each of the plurality of image patches corresponding to a processed image of the plurality of processed images comprising the RBC are stored. In an embodiment, the feature determination module 212 determines a value for the plurality of features for each of the plurality of image patches. The plurality of features may be calculating one of a value of area of the RBC in each image patch and volume of RBC m each image patch. The plurality of features may include, but are not limited to, a first feature indicative of volume of the RBC after mean centering and contrast normalization, a second feature indicative of volume of the RBC after mean centering, contrast normalization and gradient normalization in pallor, a third feature indicative of non-pallor area with continuous pallor computation (without use of any threshold), a fourth feature indicative of volume of the RBC after mean centering. The value for each feature of the plurality of features may be determined by employing known image processing techniques.

In an embodiment, the hematological parameter determination module 213 may determine the hematological parameters based on the plurality of features. Based on the value of each the plurality of features determined for each of the RBCs, the RBC may be classified as belonging to one of the pre-defined types. The plurality of features determined for each of the RBCs may be provided to a unsupervised learning model like a k-means clustering model to classify each of the RBCs into one of the predefined types. The hematological parameter determination module 213 may employ any known unsupervised learning model for classifying each the RBCs into one of the predefined types. The hematological parameter determination module 213 may identify a number of RBCs belonging to each of the predefined type. In an embodiment the hematological parameter determination module 213 may determine a percentage of RBCs belonging to each of the predefined types. Further the percentage of RBCs belonging to each of the predefined types may be provided to a regression model for determining the hematological parameters.

In an embodiment, the hematological parameters may comprise, but is not limited to, Mean Corpuscular Volume (MCV), Mean Corpuscular Hemoglobin (MCH), hemoglobin content in the blood sample. Mean Corpuscular Hemoglobin concentration (MCHC) and Hematocrit (HCT). In an embodiment, the other modules 214, may include, hut are not limited to, a report generation module.

In an embodiment, the report generation module may he used to generate a report comprising details of the hematological parameters determined by the blood analyzer 100. It may further indicate the grade of the estimation of hematological parameters.

Figure 3 and Figure 4 show an exemplary flowchart illustrating method steps 300 and 303 respectively for estimating hematological parameters in a Peripheral Blood Smear (PBS), m accordance with some embodiments of the present disclosure.

As illustrated in Figures 3 and 4, the method comprises one or more blocks for estimating hematological parameters in PBS. The method 300 and method 303 may be described in the general context of machine executable instructions. Generally, machine executable instructions can include routines, programs, objects, components, data structures, procedures, modules, and functions, which perform particular functions or implement particular abstract data types.

The order in which the method 300 and method 303 is described is not intended to be construed as a limitation, and any number of the described method blocks can be combined in any order to implement the method. Additionally, individual blocks may be deleted from the methods without departing from the scope of the subject matter described herein. Furthermore, the method can be implemented in any suitable hardware, software, firmware, or combination thereof.

At step 301, the PBS images are acquired by the image acquisition model 210, from the microscopic system 101, for processing and determining hematological parameters. The PBS images may be captured by the imaging unit of the microscopic system 101 in the monolayer region of the PBS 102. Figure 5A indicates a portion of the PBS image acquired by the image acquisition model 210. Illustrated Figure 5A is an image acquired in the monolayer region of the PBS 102. The PBS image illustrated in Figure 5A may be referred to as a first image, hereafter, in the present disclosure. In an embodiment, the image acquisition model 210 may acquire 120 images of the PBS 102. Each of the 120 images may relate to different Field of View (FOV) of the monolayer region of the PBS 102. The PBS images may he a RGB color image. The RGB color image may be realized as one of a green plane, blue plane and red plane by using suitable filters. The PBS images may be converted into a green plane form.

At step 302, the image patch extraction module 211 extracts plurality of image patches from each of the PBS images. Any known image processing technique may be employed for converting each of the PBS images into corresponding green plane format. In an embodiment, a green plane of each of the PBS images is obtained. Further, customized Otsu thresholding may be applied on the green plane of each of the PBS images to obtain the plurality of processed images. Figure SB indicates exemplary binary form of the first image. Otsu thresholding has been applied to the first image to convert it into a binary image as illustrated in Figure SB. The binary' form of the first image may be referred to as a second image hereafter in the present disclosure.

Further, plurality of image patches is extracted from each of the PBS images by the image patch extraction module 211. The image patch extraction module 211 may extract and segment plurality of image patches extracted from each of the PBS images. Each of the plurality of image patches comprises the blood cell. The blood cell may he one of RBC, WBC and platelet. The area of each of the plural ity of image patches may he pre-defmed before extraction. In an embodiment, the blood cells in the PBS, in the size range of the RBC (3-8 microns) may be considered as a patch and may be extracted. The plurality of image patches may he extracted using any existing image processing techniques used for extraction of objects from an image. In an embodiment, the image patch extraction module 211 may employ at least one of image processing techniques, pre-leamt statistical models, machine learning methods and rule-based methods or any other method which may be used for extraction of the plurality of image patches.

The extraction of the plurality of image patches may be performed using pre-tramed Artificial Intelligence models or deep learning models. The pre-trained artificial Intelligence models may be a combination of convolutional neural networks and statistical models. Each of the plurality of image patches corresponding to a binary image of the plurality of binary images comprising the RBC are stored in the image data 206. The image patch extraction module 211 is trained using the training data 205 for extraction and the parameters considered for extraction and classification. The parameters considered for extraction and classification may be one of shape of the blood cell, constituents of the blood cell, size of the blood cell and the like. In an embodiment, the image patch extraction module 211 may neglect the image patch with over lapping cells. Figure 5C indicates the RBCs identified in a PBS image (similar to the first image). The circled regions illustrated m the Figure 5C indicate the RBCs identified in the PBS image. Thus, each of the plurality of image patches comprises RBC.

At step 303, the feature determination module 212 determines the value of each feature of the plurality of features for each of the plurality of image patches. The plurality of features may include, but are not limited to, the first feature indicative of volume of the RBC after mean centering and contrast normalization, the second feature indicative of volume of the RBC after mean centering, contrast normalization and gradient normalization in pallor, the third feature indicative of non-pallor area with continuous pallor computation (without use of any threshold), the fourth feature indicative of volume of the RBC after mean centering. The value of each feature of the plurality of features may be computed by employing statistical models and known image processing techniques. In an embodiment, the plurality of features may include seven independent features like the first feature, the second feature, the third feature, the fourth feature and the area of the RBC, the pal lor ratio of the RBC and the non-pallor area of the RBC.

In an embodiment, for determining the plurality of features for each of the plurality of image patches classified as RBC, the green plane form of each of the plurality of image patches may be obtained. From the green plane form of each of the plurality of image patches intensity representations of the RBCs may be estimated. As illustrated in Figure 6A, image patch 600 indicates a green plane form of the image patch comprising a first RBC. Further, a cross sectional view 601 of the first RBC in the image patch 600 is obtained in the form of a strip of first RBC as illustrated. Further, pixel intensity graph is plotted for the pixels across the cross-sectional view 601. The pixel intensity graph is a plot of the pixel intensities in image v/s the number of pixels. Intensity graph 602 indicates the pixel intensity graph ploted for the cross-sectional view 601. Similarly, in F gure 6B image patch 603 indicates a green plane form of the image patch comprising a second RBC. Further, a cross sectional view 604 of the second RBC in the image patch 603 is obtained in the form of a strip of the second RBC as illustrated. Intensity graph 605 indicates the pixel intensity graph plotted for the cross-sectional view 604. The first RBC and the second RBC may belong to blood samples having similar volumetric parameters. A volumetric estimation of any RBC is obtained from the pixel intensity graphs. As shown in Figure 6A and Figure 6B, the pixel intensity graphs are found to be different for the first RBC and the second RBC, even though the first RBC and the second RBC have similar volumetric parameters. The variations m the pixel intensity graphs are due to staining variations and difference, distinctiveness in the colors of the first RBC and the second RBC. Due to significant variation m the appearance of the RBC in terms of color based on the type of stain applied to the blood smear, the plurality of features determined by the feature determination module 212 are intended to normalize the stain variation by using pixel intensity values.

In an embodiment, the feature determination module 212 may estimate a value of the area of the RBC, pallor ratio of the RBC and the area of a non-pallor region in the RBC. In an embodiment, RBCs have a shape similar to that of a torus (doughnut). The RBC comprises a center region called pallor and the region surrounding the pallor is termed as a non-pallor region. Figure 5D illustrates an identified RBC (after thresholding) present in an image patch. The image patch comprising the identified RBC is extracted from the second image. The center region 502 is the pallor of the identified RBC and the region 501 surrounding the region 502 is the non-pallor region of the RBC. The value of the area of RBC for the identified RBC as illustrated in Figure 51) is computed using equation 1.

Area of RBC= area of non-pallor region + area of the pallor . (1)

The value of the pallor ratio is determined using equation 2,

Pallor ratio ^::: Area of pallor/ area of RBC . (2)

The value of area of non-pallor region is determined using equation 3. Area of non-pallor region^ Area of RBC - Area of pallor (3)

Figure 7A and Figure 7B are indicative of intensity v/s frequency histograms plotted for image patch 600 (Figure 6A) and the image patch 603 (Figure 6B) respectively. The frequency which is indicated on the y-axis of the intensity v/s frequency histogram denotes the number of pixels in the image patch having a given value of intensity level (indicated on the x-axis). As seen from the Figure 7A and Figure 7B the intensity v/s frequency histograms are dissimilar due to the variations in staining and change in capture mechanism used for capturing the image patch 600 and image patch 603. As seen in Figure 7A the intensity v/s frequency histogram comprise two distinctive peaks namely a foreground peak and a background peak. The image patch 600 comprises plurality of pixels. A first set of pixels among the plurality of pixels indicate a peak value of intensity m a background region of the image patch 600. The peak indicative of the maximum frequency of intensity in the background region of the image 600 is characterized as the background peak. A second set of pixels among the plurality of pixels indicate a peak value of intensity in a foreground region of the image patch 600. The peak indicative of the maximum frequency of intensity in the foreground region of the image 600 is characterized as the foreground peak.

Figure 4 shows an exemplary flowchar illustrating method steps for estimation of the value of a first feature indicative of volume after mean centering and contrast normalization, in accordance with some embodiments of the present disclosure.

At step 401, the feature determination module 212 calculates a difference between the area of the RBC and a normalized value. The value of first feature is used for computing volume of the RBC in each of the plurality of patches. The value of the first feature is computed using the equation 4.

Area

.. (4) where the area of the RBC comprises an area of the pallor region in the RBC and the area of a non-pallor region in the RBC. The second term in the equation 4 is the normalized value winch indicates the area of pallor region. The first feature may be indicative of a first normalization technique used for normalizing the variations in contrast.

At step 402, a value of intensity of each of a plurality of pixels in each patch is determined. Further, the 212 determines the foreground peak and the background peak in the patch.

At step 403, the feature determination module 212 determines an intensity threshold valise determined based on the foreground peak and the background peak. The intensity threshold value may be an Otsu threshold. Further, the feature determination module 212 determines a difference between the value of intensity of each pixel in the patch and the intensity threshold. Considering, the Figure 7A the Otsu threshold may be centered around a pixel intensity value of 200.

At step 4Q4, the feature determination module 212 determines a distance between the foreground peak and the background peak in the patch. Considering the Figure 7A the distance between the foreground peak and the background peak may be determined.

At step 405, the feature determination module 212 determines the normalized value based on a ratio of the determined difference in step 403 and the determined distance in step 405.

The distance between the foreground peak and the background peak for the image patch 600 comprising first RBC and for image patch 603 comprising second RBC as observed in figure 6A and 6B is normalized during the computation of value of the first feature. Post the application of the first normalization technique as defined by the equation 4, the change observed in the image patch 600 and the image patch 603 is indicated in Figure 8.4. Figure 8.4 comprises an image patch 800 obtained as a result of application of the first normalization technique as defined by the equation 4 on the image patch 600. Further, Figure 8A also comprises an image patch 801 obtained as a result of application of the first normalization technique as defined by the equation 4 on the image patch 603. The difference in contrast seen between the image patch 600 and the image patch 603 is more when compared to the difference in contrast seen between the image patch 800 and the image patch 801 Thus, the difference in contrast between the image patch 600 and the image patch 603 is reduced by the use of the first normalization technique. The intensity versus frequency histogram plotted for image patch 800 and image patch 801 is as shown Figure 8B and Figure 8C respectively. On comparing Figure 7A with Figure 8B and Figure 7B with Figure 8C, it may be observed that the foreground peak and the background peak are centered at respective mean values. Mean centering of both foreground and background peaks to a particular value removes the difference in contrast between images captured from the two RBCs of similar hematological parameters. The value obtained using the equation 4 is used for computing the volume of RBC in a given patch.

In an embodiment, the feature determination module 212 determines the second feature in a manner similar to the first feature. The second feature is indicative of volume of the RBC after mean centering, contrast normalization and gradient normalization in pallor. The value of the second feature is used for computing volume of the RBC in each of the plurality of patches. The value of the second feature is computed using the equation 5.

Area

norm pixel value— threshold) /Distance)... (5) The norm pixel value is computed by a ratio of, difference in value of intensity of each of the plurality of pixels in the patch and a mean value, and a standard deviation value. The mean value is a mean of value of intensity of each of the plurality of pixels, and the standard deviation value is a standard deviation of value of intensity of each of the plurality of pixels. The norm pixel value is computed separately for the pixels in the foreground of the image patch and for the pixels in background of the image patch. The second feature is determined as per the method steps 303 with the use of norm pixel value in the equation 4 instead of the value of intensity of each of the plurality of pixels (pixel value). The second feature may be indicative of a second normalization technique used for normalizing the variations in contrast and gradient of intensities present in the pallor region of the image patch. The foreground peak and the background peak for the image patch 600 comprising first RBC and for image patch 603 comprising second RBC as observed in figure 6A and 6B is normalized individually during the computation of value of the second feature. Post the application of the second normalization technique as defined by the equation 5, the change observed in the image patch 600 and the image patch 603 is indicated in Figure 9A. Figure 9A comprises an image patch 900 obtained as a result of application of the second normalization technique as defined by the equation 5 on the image patch 600. Further, Figure 9A also comprises an image patch 901 obtained as a result of application of the second normalization technique as defined by the equation 5 on the image patch 603. The difference in contrast seen between the image patch 600 and the image patch 603 is more when compared to the difference in contrast seen between the image patch 900 and the image patch 901. For an instance, the gradient of intensities presents in the pallor region of image patch 600 and the image patch 603 is high. Thus, the pallor region in the first RBC as indicated by the image patch 400 appears to be spread out (gradient of intensities seen in the pallor region) and thus reducing the actual volume of the first RBC. Therefore, the second normalization technique clearly defines the pallor region by normalizing the intensity variations within the pallor region. The reduction in gradient of intensities within the pallor region may be clearly observed in the image 900 and image 901. The intensity versus frequency histogram plotted for image patch 900 and image patch 901 is as shown in Figure 9B and Figure 9C respectively.

On comparing Figure 7A with Figure 9B and Figure 7B with Figure 9C, it may be observed that the foreground peak and the background peak are centered at respective mean values. A distinct foreground peak and a distinct background peak may be seen m the Figure 9B and Figure 9C. Mean centering of both foreground and background peaks to a particular value removes the difference in contrast between images captured from the two RBCs of similar hematological parameters. Further, as a clear distinction may be seen between the foreground peak and the background peak (the mean and standard deviation of the foreground peak and the background peak are similar) the gradient of intensities in the pallor region of the given patch is reduced. The value obtained using the equation 5 is used for computing the volume of RBC in a given patch. In an embodiment, the feature determination module 212 determines the third feature based on a standard deviation of mean centered value of intensity of the plurality of pixels in the patch and the area of the RBC. The standard deviation of the mean centered value of intensity of the plurality of pixels in the patch is proportional to the pallor region of the RBC in the image patch. The mean centered value of intensity of each of the plurality of pixels is obtained by a ration of the value of intensity of the pixel and the intensity threshold (equation 7). The value of the third feature is determined based on the equation 6

Area of RBC (1 - Standard Deviation (Value M in RBC)) . (6)

Value M = pixel intensity value/ threshold . (7)

The value of the third feature is used for determining the actual volume of the cell. The term (Area of RBC* Standard deviation of value of pixels in RBC) in the equation 6 represents the volume of toroidal hollow (pallor region).

In an embodiment, the feature determination module 212 determines the fourth feature based on the intensity threshold. The value of the fourth feature is determined using the equation 7. The intensity threshold is the Otsu threshold. The value of the fourth feature is determined by computing a ratio of the intensity value of each of the plurality of pixels in the image patch and the intensity threshold. Further, a difference between the value obtained by the computer

Area

pixel value /threshold .. (7)

The mean of the foreground peak and the background peak as shown in each of Figure 7A and Figure 7B, are normalized using a third normalization technique realized using equation 7. Post the application of the third normalization technique, the intensity versus frequency histogram of the image patch 600 and the image patch 603 are as shown in Figure I0A and Figure 10B respectively. On comparing Figure 7A with Figure 10A and Figure 7B with Figure 10B, it may be observed that the histogram is centered at a mean intensity value of 1. The value obtained using the equation 7 is used for computing the volume of RBC in a given patch.

In an embodiment, the hematological parameter determination module 213 determines the hematological parameters based on the plurality of features. Based on the value of each the plurality of features determined for each of the RBCs, the RBC may be classified as belonging to one of the pre-defined types. The plurality of features determined for each of the RBCs may be provided to an unsupervised learning model like a k-means clustering model to classify each of the RBCs into one of the predefined types. The hematological parameter determination module 213 may employ any known unsupervised learning model for classifying each the RBCs into one of the predefined types. The pre- defined types represent major types of ceils in the entire spectrum of normal and abnormal MCV / MCH (different type of RBCs). The predefined types may be referred as clusters hereafter in the present disclosure. For instance, one cluster may represent cells with high area and Iow^r pallor ratio, a second cluster may represent cells with low area and high pallor ratio, a third cluster may represent cells with high area and high pallor ratio, and the like.

The hematological parameter determination module 213 may identify a number of RBCs belonging to each of the clusters. In an embodiment the hematological parameter determination module 213 may determine a percentage of RBCs belonging to each of the clusters. Further, the hematological parameter determination module 213 computes a histogram of cells in each of the clusters. The histogram may indicate the percentage of RBCs belonging to each of the predefined types/ clusters.

The hematological parameter determination module 213 may determine a set of independent features indicative of the percentage of RBCs belonging to each of the clusters. The set of independent features for every blood sample is computed as:

FA ^::: % of cells that are assigned to Cluster A in the blood sample.

FB = % of cells that are assigned to Cluster B in the blood sample. F_K ^:::: % of cells that are assigned to Cluster K in the blood sample.

The hematological parameter determination module 213 may determine the hematological parameters based on the set of independent features. The set of independent features types may be provided to a regression model for determining the hematological parameters. in an embodiment, the volume of RBC computed using each of the plurality of variables is used by the hematological parameter determination module 213 for determining the hematological parameters. Each of the plurality features may be used as a standalone feature of may be combined with other features for determining the hematological parameters. The performance of the blood analyzer 100, was analyzed, and blood analyzer 100 is validated on a set of 140 samples. Out of the 140 samples, 90 are prepared using MGG stain and 50 are prepared using Leishman stain. The hematological parameters estimated using the plurality of features are compared with a ground truth. The ground truth defines a value of the hematological parameter estimated using a known technique. The Table 1 indicates the correlation between the hematological parameters determined using the plurality of features and the ground truth. The area of RBC as indicated by equation 1 may be referred as a fifth feature (F5). The pallor ratio as indicated by equation 2 may be referred as a sixth feature (F6) and the area of non-pallor region as indicated by equation 3 may be referred as a seventh feature (F7). The first feature, second feature, the third feature and the fourth feature may be referred as FI, F2, F3 and F4 respectively.

The values indicated in Table 1 are Root Mean Square Error (RMSE) values. The RMSE value indicates the error present in the hematological parameter estimated by using the plurality of features and the ground truth. The lesser the value of RMSE indicates a better correlation between the determined hematological parameter using the plurality of features and the ground truth. As indicated by Table 1 , the RMSE value is the least when all the features Fl to F7 are used.

Computer System

Figure 11 illustrates a block diagram of an exemplary computer system 1000 for implementing embodiments consistent with the present disclosure. In an embodiment, the computer system 1000 is used to implement the blood analyzer 100. The computer system 1000 may comprise a central processing unit (“CPU” or“processor”) 1002. The processor 1002 may comprise at least one data processor for executing program components for determining hematological parameters m the PBS 102. The processor 1002 may include specialized processing units such as integrated system (bus) controllers, memory management control units, floating point units, graphics processing units, digital signal processing units, etc.

The processor 1002 may be disposed in communication with one or more mput/output (I/O) devices (not shown) via TO interface 1001. The TO interface 1001 may employ communication protocols/methods such as, without limitation, audio, analog, digital, monoaural, RCA, stereo, IEEE- 1394, serial bus, universal serial bus (USB), infrared, PS/2, BNC, coaxial, component, composite, digital visual interface (DVI), high- definition multimedia interface (HDMI), RF antennas, S-Video, VGA, IEEE 802. n /b/g/n/x, Bluetooth, cellular (e.g., code-division multiple access (CDMA), high-speed packet access (HSPA+), global system for mobile communications (GSM), long-term evolution (LTE), WIMax, or the like), etc. Using the I/O interface 1001, the computer system 1000 may communicate with one or more I/O devices. For example, the input device 1010 may be an antenna, keyboard, mouse, joystick, (infrared) remote control, camera, card reader, fax machine, dongle, biometric reader, microphone, touch screen, touchpad, trackball, stylus, scanner, storage device, transceiver, video device/source, etc. In an embodiment, the input device 1010 may be the microscopic system 101. The output device 1011 may be a printer, fax machine, video display (e.g., cathode ray tube (CRT), liquid crystal display (LCD), light-emitting diode (LED), plasma, Plasma display panel (PDP), Organic light-emitting diode display (OLED) or the like), audio speaker, etc.

In some embodiments, the computer system 1000 is connected to a server 1012. through a communication network 1009. The server 1012 may implement image processing tools used by the computer system 1000. The processor 1002 may be disposed in communication with the communication network 1009 via a network interface 1003. The network interface 1003 may communicate with the communication network 1009. The network interface 1003 may employ connection protocols including, without limitation, direct connect, Ethernet (e.g., twisted pair 10/100/1000 Base T), transmission control protocol/internet protocol (TCP/IP), token ring, IEEE 802.1 1 a/b/g/n/x, etc. The communication network 1009 may include, without limitation, a direct interconnection, local area network (LAN), wide area network (WAN), wireless network (e.g., using Wireless Application Protocol), the Internet, etc. Using the network interface 1003 and the communication network 1009, the computer system 1000 may communicate with the classifier model 1012, The network interface 1003 may employ connection protocols include, but not limited to, direct connect, Ethernet (e.g., twisted pair 10/100/1000 Base T), transmission control protocol/internet protocol (TCP/IP), token ring, IEEE 802.1 1 a/b/g/n/x, etc.

The communication network 1009 includes, but is not limited to, a direct interconnection, an e-commerce network, a peer to peer (P2P) network, local area network (LAN), wide area network (WAN), wireless network (e.g., using Wireless Application Protocol), the Internet, Wi-Fi and such. The first network and the second network may either be a dedicated network or a shared network, which represents an association of the different types of networks that use a variety of protocols, for example. Hypertext Transfer Protocol (HTTP), Transmission Control Protocol/Internet Protocol (TCP/IP), Wireless Application Protocol (WAP), etc., to communicate with each other. Further, the first network and the second network may include a variety of network devices, including routers, bridges, servers, computing devices, storage devices, etc. in some embodiments, the processor 1002 may be disposed m communication with a memory 1005 (e.g., RAM, ROM, etc. not shown in figure 5) via a storage interface 1004. The storage interface 1004 may connect to memory' 1005 including, without limitation, memory drives, removable disc drives, etc., employing connection protocols such as serial advanced technology attachment (SATA), Integrated Drive Electronics (IDE), IEEE- 1394, Universal Serial Bus (USB), fiber channel, Small Computer Systems Interface (SCSI), etc. The memory drives may further include a drum, magnetic disc drive, magneto-optical drive, optical drive, Redundant Array of Independent Discs (RAID), solid-state memory devices, solid-state drives, etc.

The memory 1005 may store a collection of program or database components, including, without limitation, user interface 1006, an operating system 1007, web server 1008 etc. In some embodiments, computer system 1000 may store user/application data1006, such as, the data, variables, records, etc., as described in this disclosure. Such databases may be implemented as fault-tolerant, relational, scalable, secure databases such as Oracle ® or Sybase®.

The operating system 1007 may facilitate resource management and operation of the computer system 1000. Examples of operating systems include, without limitation, APPLE MACINTOSH¹¹ OS X, UNIX^R, UNIX-like system distributions (EG., BERKELEY SOFTWARE DISTRIBUTION™ (BSD), FREEBSD™¹, NETBSD™, OPENBSD™, etc ), LINUX DISTRIBUTIONS™¹ (E.G., RED HAT™¹, IJBUNTU™¹, KUBUNTU™, etc.), IBM™¹ OS/2, MICROSOFT™¹ WINDOWS™¹ (XP™, VISTA™/7/8, 10 etc,), APPLE™ !OS™. GOOGLE™ ANDROID™¹, BLACKBERRY™ OS, or the like. In some embodiments, the computer system 1000 may implement a web browser 1008 stored program component. The web browser 1008 may be a hypertext viewing application, for example MICROSOFT^® INTERNET EXPLORER™, GOOGLE^® CHROME™⁰, MOZILLA^® FIREFQX™, APPLE^® SAFARI™, etc, Secure web browsing may be provided using Secure Hypertext Transport Protocol (HTTPS), Secure Sockets Layer (SSL), Transport Layer Security (TLS), etc. Web browsers 1008 may utilize facilities such as AJAX™, DHTML™, ADOBE^® FLASH™, JAVASCRIPT™, JAVA™, Application Programming Interfaces (APIs), etc. In some embodiments, the computer system 1000 may implement a mail server stored program component. The mail server may be an Internet mail server such as Microsoft Exchange, or the like. The mail server may utilize facilities such as ASP™, ACTIVEX™, ANSI^1®1 C++/C#, MICROSOFT^®, .NET™, CGI SCRIPTS™, JAVA™, JAVASCRIPT™, PERL™, PHP™, PYTHON™¹, WEBOBJECTS™, etc. The mail server may utilize communication protocols such as Internet Message Access Protocol (IMAP), Messaging Application Programming Interface (MAPI), MICROSOFT^® exchange, Post Office Protocol (POP), Simple Mail Transfer Protocol (SMTP), or the like. In some embodiments, the computer system 1000 may implement a mail client stored program component. The mail client may be a mail viewing application, such as APPLE^® MAIL™, MICROSOFT^® ENTOURAGE™¹, MICROSOFT^® OUTLOOK™¹, MOZILLA^® THUNDERBIRD™, etc.

Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term“computer-readable medium” should be understood to include tangible items and exclude carrier waves and transient signals, i.e., be non-transitory. Examples include Random Access Memory (RAM), Read-Only Memory (ROM), volatile memory, nonvolatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media. Advantages of the embodiment of the present disclosure are illustrated herein.

Embodiments of the present disclos ure relate to a method and system for estimating the hematological parameters in the PBS. The system acquires the plurality of images from the monolayer region of the PBS, thereby producing an unbiased estimation of the hematological parameters.

The method and system are proficient and robust in estimating hematological parameters efficiently. The information of the plurality of features when appended with the information of RBC and non-RBC cells, helps in building robust set of parameters which are stain agnostic and works well for cases of overlapping cells in images.

The method and system is smear agnostic. The system is robust and proficient m estimating the hematological parameters even when different image capturing devices are used to capture images of the PBS.

The described operations may be implemented as a method, system or article of manufacture using standard programming and/or engineering techniques to produce software, firmware, hardware, or any combination thereof. The described operations may be implemented as code maintained in a“non-transitory computer readable medium”, where a processor may read and execute the code from the computer readable medium. The processor is at least one of a microprocessor and a processor capable of processing and executing the queries. A non-transitory computer readable medium may comprise media such as magnetic storage medium (e.g., hard disk drives, floppy disks, tape, etc.), optical storage (CD-ROMs, DVDs, optical disks, etc.), volatile and non-volatile memory devices (e.g., EEPROMs, ROMs, PROMs, RAMs, DRAMs, SRAMs, Flash Memory, firmware, programmable logic, etc.), etc. Further, non-transitory computer-readable media comprise all computer-readable media except for a transitory. The code implementing the described operations may further be implemented in hardware logic (e.g., an integrated circuit chip, Programmable Gate Array (PGA), Application Specific Integrated Circuit (ASIC), etc.). Still further, the code implementing the described operations may be implemented in“transmission signals”, where transmission signals may propagate through space or through a transmission media, such as an optical fiber, copper wire, etc. The transmission signals in which the code or logic is encoded may further comprise a wireless signal, satellite transmission, radio waves, infrared signals, Bluetooth, etc. The transmission signals m which the code or logic is encoded is capable of being transmitted by a transmitting station and received by a receiving station, where the code or logic encoded in the transmission signal may be decoded and stored in hardw¾re or a non-transitory computer readable medium at the receiving and transmitting stations or devices. An“article of manufacture” comprises non-transitory computer readable medium, hardware logic, and/or transmission signals in winch code may be implemented. A device in which the code implementing the described embodiments of operations is encoded may comprise a computer readable medium or hardware logic. Of course, those skilled in the art will recognize that many modifications may be made to this configuration without departing from the scope of the invention, and that the article of manufacture may comprise suitable information bearing medium known in the art.

The terms“an embodiment”,“embodiment”,“embodiments”,“the embodiment”, “the embodiments”, “one or more embodiments”, “some embodiments”, and “one embodiment” mean“one or more (but not all) embodiments of the invention(s)” unless expressly specified otherwise.

The terms “including”, “comprising”, “having” and variations thereof mean “including but not limited to”, unless expressly specified otherwise.

The enumerated listing of items does not imply that any or all of the items are mutually exclusive, unless expressly specified otherwise.

The terms“a”,“an” and“the” mean“one or more”, unless expressly specified otherwise.

A description of an embodiment with several components in communication with each other does not imply that all such components are required. On the contrary a variety of optional components are described to illustrate the wide variety of possible embodiments of the invention.

When a single device or article is described herein, it will be readily apparent that more than one device/article (whether or not they cooperate) may be used in place of a single device/article. Similarly, where more than one device or article is described herein (whether or not they cooperate), it will be readily apparent that a single device/article may be used m place of the more than one device or article or a different number of devices/articles may be used instead of the shown number of devices or programs. The functionality and/or the features of a device may be alternatively embodied by one or more other devices which are not explicitly described as having such functionality/features. Thus, other embodiments of the invention need not include the device itself.

The illustrated operations of Figure 3 and Figure 4 show certain events occurring in a certain order. In alternative embodiments, certain operations may be performed in a different order, modified or removed. Moreover, steps may be added to the above described logic and still conform to the described embodiments. Further, operations described herein may occur sequentially or certain operations may be processed in parallel. Yet further, operations may be performed by a single processing unit or by distributed processing units.

Finally, the language used in the specification has been principally selected for readability and instructional purposes, and it may not have been selected to delineate or circumscribe the inventive subject matter. It is therefore intended that the scope of the invention be limited not by this detailed description, but rather by any claims that issue on an application based here on. Accordingly, the disclosure of the embodiments of the invention is intended to be illustrative, but not limiting, of the scope of the invention, which is set forth in the following claims.

While various aspects and embodiments have been disclosed herein, other aspects and embodiments will be apparent to those skilled in the art. The various aspects and embodiments disclosed herein are for purposes of illustration and are not intended to be limiting, with the true scope and spirit being indicated by the following claims. Referral Numerals:

Claims

We claim:

1. A method for determining hematological parameters in a blood smear, comprising: receiving, by a blood analyzer, a plurality of images of a blood smear from an image capturing unit configured to focus and capture images of the blood smear, wherein the plurality of images is captured from a monolayer of the blood smear, wherein each image is processed for extracting a plurality of patches, wherein each patch comprises a Red Blood Cell (RBC);

determining, by the blood analyzer, a plurality of features required for calculating volume of the RBC in each patch, wherein a value of a first feature from the plurality of features is determined by calculating a difference between an area of the RBC and a normalized value, wherein the area of the RBC comprises an area of a pallor region in the RBC and an area of a non-pallor region in the RBC, and the normalized value indicates the area of pallor region, wherein computing the normalized value for each patch comprises:

determining, a value of intensity of each of a plurality of pixels in each patch, wherein a first set of pixels among the plurality of pixels indicates a peak value of intensity in a background region and is characterized as background peak, and a second set of pixels among the plurality of pixels indicate a peak value of intensity in a foreground region and is characterized as foreground peak; determining, a difference between the value of intensity of each pixel in the patch and an intensity threshold value determined based on the foreground peak and the background peak;

determining, a distance between the foreground peak and the background peak in the patch; and

determining, the normalized value based on a ratio of the determined difference and the distance;

wherein the hematological parameters are determined based on the plurality of features.

2. The method as claimed in claim l, wherein determining hematological parameters comprises: classifying each RBC into predefined types based on area of the RBC, area of pallor region and area of the non-pallor region;

identifying a number of RJBCs belonging to each of the predefined type; and determining the hematological parameters based on the number of RBCs of each type;

3. The method as claimed in claim 1, wherein the hematological parameters comprises at least one of Mean Corpuscular Volume (MCV), Mean Corpuscular Hemoglobin (MCH), hemoglobin content m a blood sample, Mean Corpuscular Hemoglobin concentration (MCHC) and Hematocrit (HCT).

4. The method as claimed in claim 1, wherein a value of a second feature from the plurality of features is determined by calculating a difference between the area of the RBC and a first normalized value, wherein the first normalized value is computed using the value of intensity of each of the plurality of pixels m each patch which is characterized by a ratio of, difference in value of intensity of each of the plurality of pixels and a mean value, and a standard deviation value, wherein the mean value is a mean of value of intensity of each of the plural ity of pixels, and the standard deviati on value is a standard deviation of value of intensity of each of the plurality of pixels.

5. The method as claimed in claim 4, wherein a third feature from the plurality of features is determined based on the standard deviation of the value of intensity of the plurality of pixels in the patch and the area of the RBC.

6. A blood analyzer for determining hematological parameters in a blood smear, said blood analyzer comprising:

a processor; and

a memory, communicatively coupled with the processor, storing processor executable instructions, which, on execution causes the processor to:

receive, a plurality of images of a blood smear from an image capturing unit configured to focus and capture images of the blood smear, wherein the plurality of images is captured from a monolayer of the blood smear, wherein each image is processed for extracting a plurality of patches, wherein each patch comprises a Red Blood Cell (RBC);

determine, a plurality of features required for calculating volume of the RBC in each patch, wherein a value of a first feature from the plurality of features is determined by calculating a difference between an area of the RBC and a normalized value, wherein the area of the RBC comprises an area of a pallor region in the RBC and an area of a non-pallor region in the RBC, and the normalized value indicates the area of pallor region, wherein computing the normalized value for each patch comprises:

determining a value of intensity of each of a plurality of pixels in each patch, wherein a first set of pixels among the plurality of pixels indicates a peak value of intensity in a background region and is characterized as background peak, and a second set of pixels among the plurality of pixels indicate a peak value of intensity in a foreground region and is characterized as foreground peak;

determining a difference between the value of intensity of each pixel in the patch and an intensity threshold value determined based on the foreground peak and the background peak;

determining, the normali zed value based on a ratio of the determin ed difference and the distance;

7. The blood analyzer as claimed m claim 6, wherein determining hematological parameters comprises:

classifying each RBC into predefined types based on area of the RBC, area of pallor region and area of the non-pallor region;

identifying a number of RBCs belonging to each of the predefined type; and determining the hematological parameters based on the number of RBCs of each type;

8. The blood analyzer as claimed in claim 6, wherein the hematological parameters comprises at least one of Mean Corpuscular Volume (MCV), Mean Corpuscular Hemoglobin (MCH), hemoglobin content in a blood sample. Mean Corpuscular Hemoglobin concentration (MCHC) and Hematocrit (HCT).

9. The blood analyzer as claimed in claim 6, wherein a value of a second feature from the plurality of features is determined by calculating a difference between the area of the RBC and a first normalized value, wherein the first normalized value is computed using the value of intensity of each of the plurality of pixels m each patch which is characterized by a ratio of, difference in value of intensity of each of the plurality of pixels and a mean value, and a standard deviation value, wherein the mean value is a mean of value of intensity of each of the plurality of pixels, and the standard deviation value is a standard deviation of value of intensity of each of the plurality of pixels.

10. The blood analyzer as claimed in claim 6, wherein a third feature from the plurality of features is determined based on the standard deviation of the value of intensity of the plural ity of pixels in the patch and the area of the RBC.