Vishnu Vardhan Makkapati

Segmentation of malaria parasites in peripheral blood smear images

Detection of malaria parasites in stained blood smears is critical for treatment of the disease. Automation of this process will help in reducing the time taken for diagnosis and the chance for human errors. However, the variability and artifacts in microscope images of blood samples pose significant challenges for accurate detection. A scheme based on HSV color space that segments Red Blood Cells and parasites by detecting dominant hue range and by calculating optimal saturation thresholds is presented in this paper. Methods that are less computation-intensive than existing approaches are proposed to remove artifacts. The scheme is evaluated using images taken from Leishman-stained blood smears. Sensitivity and specificity of the scheme are found to be 83% and 98% respectively.

Segmentation and classification of tuberculosis bacilli from ZN-stained sputum smear images

Quality of tuberculosis (TB) diagnosis by manual observation varies depending on the quality of the smear and skill of the pathologist. To overcome this problem, a method for diagnosis of TB from ZN-stained sputum smear images is presented in this paper. Hue color component based approach is proposed to segment the bacilli by adaptive choice of the hue range. The bacilli are declared to be valid or invalid depending on the presence of beaded structure inside them. The beaded structure is segmented by thresholding the saturation component of the bacilli pixels. Clumps of bacilli and other artifacts are removed by thresholding the area, thread length and thread width parameters of the bacilli. Results presented for several images taken from different patients show that the scheme detects the presence of TB accurately.

Improved wavelet-based microscope autofocusing for blood smears by using segmentation

Video-based autofocus has become a viable option for microscopes due to the availability of fast microcomputers and cameras that provide high frame rates. The methods proposed plot a measure of the focus vs. the frame number, commonly referred to as focus function which results in a peak when the in focus frame is reached. Recently, generic waveletbased schemes have been proposed that offer varying degrees of performance depending on the specimens being observed. The performance of these methods can be improved if the nature of the specimen being observed is known. One such scheme for blood smears based on segmentation is presented in this paper. It exploits the fact that the primary objects of interest, the Red Blood Cells (RBC), have a smooth texture. It segments the RBCs and then applies the wavelet-based focus measure. This results in a smooth focus function which permits accurate detection of the in focus frame. The proposed scheme is evaluated using several videos taken from blood smears and the results show that segmentation step improves the wavelet-based measure.

Clump splitting based on detection of dominant points from contours

The presence of clumps in biological cell images may degrade the performance of automated disease detection methods using them. We present techniques to split the clumps based on dominant point detection from contours. The selection of optimal dominant points and split lines is achieved using some logical rules based on geometry of the clumps. The scheme first discards some of the dominant points by processing three successive dominant points and labels the remaining points as split points. The split points are then joined by finding pairs of optimal split points that achieve a good split. Pairs of split points that are opposite to each other are joined first and then the remaining split points are joined by using appropriate heuristics. The performance of the scheme is evaluated using several blood smear images and the results show that the method is capable of handling complex clumps.

Ontology-based malaria parasite stage and species identification from peripheral blood smear images

The diagnosis and treatment of malaria infection requires detecting the presence of the malaria parasite in the patient as well as identification of the parasite species. We present an image processing-based approach to detect parasites in microscope images of a blood smear and an ontology-based classification of the stage of the parasite for identifying the species of infection. This approach is patterned after the diagnosis approach adopted by a pathologist for visual examination, and hence, is expected to deliver similar results. We formulate several rules based on the morphology of the basic components of a parasite, namely, chromatin dot(s) and cytoplasm, to identify the parasite stage and species. Numerical results are presented for data taken from various patients. A sensitivity of 88% and a specificity of 95% is reported by evaluation of the scheme on 55 images.

Motion-based segmentation of chest and abdomen region of neonates from videos

Respiration rate (RR) is one of the important vital signs used for clinical monitoring of neonates in intensive care units. Due to the fragile skin of the neonates, it is preferable to have monitoring systems with minimal contact with the neonate. Recently, several methods have been proposed for contact-free monitoring of vital signs using a video camera. Detection of the chest-and-abdomen region of the neonate is crucial to determining the respiration rate accurately. We propose a technique for automatic selection of the region of interest (ROI) in neonates using motion. Our approach is based on the observation that points on the chest-and-abdomen region, and hence, the corresponding optic flow vectors, exhibit coherency in the motion caused by breathing. The motion induced due to the movement of the neonate (e.g., hands and legs) is not coherent and hence does not exhibit the characteristics of respiratory motion. We evaluate the proposed technique using several videos of neonates and demonstrate that it picks up the ROI accurately in spite of the movement of the neonate. We compare its performance with that of the standard motion history image (MHI) framework, using different metrics. Results indicate that our method can be profitably employed in RR studies.

Extreme Compression of Weather Radar Data

A method for achieving extreme levels of compression of high-volume weather radar data is presented. Weather reflectivity contours, as per National Weather Service or custom thresholds, are processed by tracing their departure from a smoothed version to obtain the local extrema which serve as control points. The control points, which are transmitted in relative coordinates for further compression, are interpolated using a second-degree B-spline to retrieve the contours. The encoding-decoding method is capable of capturing the random undulations inherent in weather contours. It is shown that over two orders of magnitude of compression is possible without perceptible loss of meteorological information. Multiple enhancements to the basic method are quantitatively studied and compared with the existing methods for radar data compression.

Camera based estimation of respiration rate by analyzing shape and size variation of structured light

Respiration rate is a key parameter that is monitored in intensive care units. The current solutions for respiration rate require that a sensor is placed in contact with the subject to derive it. However, this will cause discomfort to the subject and may damage the fragile skin if the subject were a neonate. There are several other applications where the respiration signal is used in clinical settings. The respiration signal is used to correct for organ motion during diagnosis scans, e.g., computed tomography. It is expected that the respiration signal results in clear peaks and troughs during the breathing cycle of the subject so that these corrections can be made. We propose a contactless system and method by using a camera to derive a faithful respiration signal. We project a circular dot of light onto the chest and abdomen region of the subject and monitor the shape and size changes of it using a camera. In an oblique projection of the structured light, both the shape and size of the dot will vary with breathing. We segment the dot and derive the respiration signal from it. Numerical results presented show that we are able to obtain the respiration rate accurately.

Computation of tortuosity of two dimensional vessels

Vessel tortuosity is an important parameter in determining the presence or the severity of various diseases that are diagnosed based on the pathology of vessels. An accurate quantification of vessel tortuosity would therefore be of great clinical significance. Usually, the branches of blood vessels are also captured, along with the major blood vessel, in the medical images, thereby posing a challenge to the computation of tortuosity. This necessitates some pre-processing of the images, to remove the branches of blood vessels, accomplished using contour tracing techniques, as proposed in this paper. In this paper, we propose a novel metric to compute tortuosity for two dimensional vessels, with brief indications of its implementation using image processing techniques, and provide evidence of its advantages over other methods.

Segmentation based microscope autofocusing for blood smears

Focusing is a critical step in microscope observation of slides. Autoscanning microscopes have to perform autofocus function accurately to record high quality images that may be later analyzed using sophisticated algorithms. Video based autofocus has become a viable option due to the availability of high computing power and cameras that provide high resolution images. A focus function which obtains a peak value when an image in focus has been encountered is used by these methods. In this paper a novel focus function based on the shape of the objects being observed is proposed. A segmentation based approach to autofocus blood smears where the primary objects being observed, the red blood cells (RBC), are circular is presented. The scheme first segments the RBCs and then determines the valid RBCs by using area criterion. The average form factor and eccentricity values of the valid RBCs in a given frame are computed. A plot of these parameters vs. the frame number will result in a peak and trough respectively for the in focus image. Results presented for various data sets show that form factor is a suitable autofocus function.