IDRiD: Diabetic Retinopathy–Segmentation and Grading Challenge

Diabetic Retinopathy (DR) is the most common cause of avoidable vision loss, predominantly affecting the working-age population across the globe. Screening for DR, coupled with timely consultation and treatment, is a globally trusted policy to avoid vision loss. However, implementation of DR screening programs is challenging due to the scarcity of medical professionals able to screen a growing global diabetic population at risk for DR. Computer-aided disease diagnosis in retinal image analysis could provide a sustainable approach for such large-scale screening effort. The recent scientific advances in computing capacity and machine learning approaches provide an avenue for biomedical scientists to reach this goal. Aiming to advance the state-of-the-art in automatic DR diagnosis, a grand challenge on “Diabetic Retinopathy – Segmentation and Grading” was organized in conjunction with the IEEE …

Prasanna Porwal, Samiksha Pachade, Manesh Kokare, Debdoot Sheet, Oindrila Saha, Rachana Sathish et al.

Medical image analysis

Adversarially Trained Deep Neural Semantic Hashing Scheme for Subjective Search in Fashion Inventory

The simple approach of retrieving a closest match of a query image from one in the gallery, compares an image pair using sum of absolute difference in pixel or feature space. The process is computationally expensive, ill-posed to illumination, background composition, pose variation, as well as inefficient to be deployed on gallery sets with more than 1000 elements. Hashing is a faster alternative which involves representing images in reduced dimensional simple feature spaces. Encoding images into binary hash codes enables similarity comparison in an image-pair using the Hamming distance measure. The challenge, however, lies in encoding the images using a semantic hashing scheme that lets subjective neighbors lie within the tolerable Hamming radius. This work presents a solution employing adversarial learning of a deep neural semantic hashing network for fashion inventory retrieval. It consists of a feature extracting convolutional neural network (CNN) learned to (i) minimize error in classifying type of clothing,(ii) minimize hamming distance between semantic neighbors and maximize distance between semantically dissimilar images,(iii) maximally scramble a discriminator’s ability to identify the corresponding hash code-image pair when processing a semantically similar query-gallery image pair. Experimental validation for fashion inventory search yields a mean average precision (mAP) of 90.65% in finding the closest match as compared to 53.26% obtained by the prior art of deep Cauchy hashing for hamming space retrieval.

Saket Singh, Debdoot Sheet, Mithun Dasgupta

2019 4th International Workshop on Fashion and KDD, 25th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

Multitask Learning of Temporal Connectionism in Convolutional Networks using a Joint Distribution Loss Function to Simultaneously Identify Tools and Phase in Surgical Videos

Surgical workflow analysis is of importance for understanding onset and persistence of surgical phases and individual tool usage across surgery and in each phase. It is beneficial for clinical quality control and to hospital administrators for understanding surgery planning. Video acquired during surgery typically can be leveraged for this task. Currently, a combination of convolutional neural network (CNN) and recurrent neural networks (RNN) are popularly used for video analysis in general, not only being restricted to surgical videos. In this paper, we propose a multi-task learning framework using CNN followed by a bi-directional long short term memory (Bi-LSTM) to learn to encapsulate both forward and backward temporal dependencies. Further, the joint distribution indicating set of tools associated with a phase is used as an additional loss during learning to correct for their co-occurrence in any predictions. Experimental evaluation is performed using the Cholec80 dataset. We report a mean average precision (mAP) score of 0.99 and 0.86 for tool and phase identification respectively which are higher compared to prior-art in the fie

Shanka Subhra Mondal, Rachana Sathish, Debdoot Sheet

MedImage Workshop, 2018 Indian Conference on Vision, Graphics and Image Processing

Learning with Multitask Adversaries using Weakly Labelled Data for Semantic Segmentation in Retinal Images

A prime challenge in building data driven inference models is the unavailability of statistically significant amount of labelled data. Since datasets are typically designed for a specific purpose, and accordingly are weakly labelled for only a single class instead of being exhaustively annotated. Despite there being multiple datasets which cumulatively represents a large corpus, their weak labelling poses challenge for direct use. As in case of retinal images which have inspired development of data driven learning based algorithms for segmenting anatomical landmarks like vessels and optic disc as well as pathologies like microaneurysms, hemorrhages, hard exudates and soft exudates; aspired to segment all using only a single fully convolutional neural network (FCN), there is no single dataset with all classes annotated. We solve this problem by training a single network using separate weakly labelled datasets. Essentially we use an adversarial learning approach over a semantic segmentation FCN, where the objectives of discriminators are to learn to (a) predict which of the classes are actually present in the input fundus image, and (b) distinguish between manual annotations vs. segmented results for each of the classes. The first discriminator works to enforce the network to segment those classes which are present in the fundus image although may not have been annotated ie all retinal images have vessels while pathology datasets may not have annotated them in the dataset. The second discriminator contributes to making the segmentation result as realistic as possible. We experimentally demonstrate using weakly labelled datasets of DRIVE …

Oindrila Saha, Rachana Sathish, Debdoot Sheet

International Conference on Medical Imaging with Deep Learning

Significance of Residual Learning and Boundary Weighted Loss in Ischaemic Stroke Lesion Segmentation

Radiologists use various imaging modalities to aid in different tasks like diagnosis of disease, lesion visualization, surgical planning and prognostic evaluation. Most of these tasks rely on the the accurate delineation of the anatomical morphology of the organ, lesion or tumor. Deep learning frameworks can be designed to facilitate automated delineation of the region of interest in such cases with high accuracy. Performance of such automated frameworks for medical image segmentation can be improved with efficient integration of information from multiple modalities aided by suitable learning strategies. In this direction, we show the effectiveness of residual network trained adversarially in addition to a boundary weighted loss. The proposed methodology is experimentally verified on the SPES-ISLES 2015 dataset for ischaemic stroke segmentation with an average Dice coefficient of 0.881 for penumbra and 0.877 for core. It was observed that addition of residual connections and boundary weighted loss improved the performance significantly.

Ronnie Rajan, Rachana Sathish, Debdoot Sheet

2019 International Conference on Medical Imaging with Deep Learning

SUMNet: Fully Convolutional Model for Fast Segmentation of Anatomical Structures in Ultrasound Volumes

Ultrasound imaging is generally employed for real-time investigation of internal anatomy of the human body for disease identification. Delineation of the anatomical boundary of organs and pathological lesions is quite challenging due to the stochastic nature of speckle intensity in the images, which also introduces visual fatigue for the observer. This paper introduces a fully convolutional neural network based method to segment organ and pathologies in ultrasound volume by learning the spatial-relationship between closely related classes in the presence of stochastically varying speckle intensity. We propose a convolutional encoder-decoder like framework with (i) feature concatenation across matched layers in encoder and decoder and (ii) index passing based unpooling at the decoder for semantic segmentation of ultrasound volumes. We have experimentally evaluated the performance on publicly available datasets consisting of 10 intravascular ultrasound pullback acquired at 20MHz and 16freehand thyroid ultrasound volumes acquired 11-16 MHz. We have obtained a dice score of0.93±0.08 and 0.92±0.06 respectively, following a 10 -fold cross-validation experiment while processing frame of 256x384 pixel in 0.035 s and a volume of 256x384x384 voxel in 13.44 s.

Sumanth Nandamuri, Debarghya China, Pabitra Mitra, Debdoot Sheet

2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019)

Learning a Deep Convolution Network with Turing Test Adversaries for Microscopy Image Super Resolution

Adversarially trained deep neural networks have significantly improved performance of single image super resolution, by hallucinating photorealistic local textures, thereby greatly reducing the perception difference between a real high resolution image and its super resolved (SR) counterpart. However, application to medical imaging requires preservation of diagnostically relevant features while refraining from introducing any diagnostically confusing artifacts. We propose using a deep convolutional super resolution network (SRNet) trained for (i) minimising reconstruction loss between the real and SR images, and (ii) maximally confusing learned relativistic visual Turing test (rVTT) networks to discriminate between (a) pair of real and SR images (T1) and (b) pair of patches in real and SR selected from region of interest (T2). The adversarial loss of T1 and T2 while backpropagated through SRNet helps it learn to reconstruct pathorealism in the regions of interest such as white blood cells (WBC) in peripheral blood smears or epithelial cells in histopathology of cancerous biopsy tissues, which are experimentally demonstrated here. Experiments performed for measuring signal distortion loss using peak signal to noise ratio (pSNR) and structural similarity (SSIM) with variation of SR scale factors, impact of rVTT adversarial losses, and impact on reporting using SR on a commercially available artificial intelligence (AI) digital pathology system substantiate our claims.

Francis Tom, Himanshu Sharma, Dheeraj Mundhra, Tathagato Rai Dastidar, Debdoot Sheet

2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019)

UltraCompression: Framework for High Density Compression of Ultrasound Volumes using Physics Modeling Deep Neural Networks

Ultrasound image compression by preserving speckle-based key information is a challenging task. In this paper, we introduce an ultrasound image compression framework with the ability to retain realism of speckle appearance despite achieving very high-density compression factors. The compressor employs a tissue segmentation method, transmitting segments along with transducer frequency, number of samples and image size as essential information required for decompression. The decompressor is based on a convolutional network trained to generate patho-realistic ultrasound images which convey essential information pertinent to tissue pathology visible in the images. We demonstrate generalizability of the building blocks using two variants to build the compressor. We have evaluated the quality of decompressed images using distortion losses as well as perception loss and compared it with other off the shelf solutions. The proposed method achieves a compression ratio of725:1 while preserving the statistical distribution of speckles. This enables image segmentation on decompressed images to achieve dice score of 0.89±0.11 , which evidently is not so accurately achievable when images are compressed with current standards like JPEG, JPEG 2000, WebP and BPG. We envision this frame work to serve as a roadmap for speckle image compression standards.

Debarghya China, Francis Tom, Sumanth Nandamuri, Aupendu Kar, Mukundhan Srinivasan, Pabitra Mitra, Debdoot Sheet

2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019)

Fully Convolutional Model for Variable Bit Length and Lossy High Density Compression of Mammograms

Early works on medical image compression date to the 1980’s with the impetus on deployment of teleradiology systems for high-resolution digital X-ray detectors. Commercially deployed systems during the period could compress 4,096× 4,096 sized images at 12 bpp to 2 bpp using lossless arithmetic coding, and over the years JPEG and JPEG2000 were imbibed reaching upto 0.1 bpp. Inspired by the reprise of deep learning based compression for natural images over the last two years, we propose a fully convolutional autoencoder for diagnostically relevant feature preserving lossy compression. This is followed by leveraging arithmetic coding for encapsulating high redundancy of features for further high-density code packing leading to variable bit length. We demonstrate performance on two different publicly available digital mammography datasets using peak signal-to-noise ratio (pSNR), structural similarity (SSIM) index and domain adaptability tests between datasets. At high density compression factors of> 300×(0.04 bpp), our approach rivals JPEG and JPEG2000 as evaluated through a Radiologist’s visual Turing test.

Aupendu Kar, Sri Phani Krishna Karri, Nirmalya Ghosh, Ramanathan Sethuraman, Debdoot Sheet

The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops

Full list of publications

(You may also visit the Google Scholar page)

  • 2015:

    1. Debapriya Maji, Anirban Santara, Sambuddha Ghosh, Debdoot Sheet, Pabitra Mitra, "Deep neural network and random forest hybrid architecture for learning to detect retinal vessels in fundus images", 2015 37th annual international conference of the IEEE Engineering in Medicine and Biology Society (EMBC) pp. 3029-3032

    2. Debdoot Sheet, Sri Phani Krishna Karri, Amin Katouzian, Nassir Navab, Ajoy K Ray, Jyotirmoy Chatterjee, "Deep learning of tissue specific speckle representations in optical coherence tomography and deeper exploration for in situ histology", 2015 IEEE 12th International Symposium on Biomedical Imaging (ISBI) pp. 777-780

    3. Abhijit Guha Roy, Debdoot Sheet, "Dasa: Domain adaptation in stacked autoencoders using systematic dropout", 2015 3rd IAPR Asian Conference on Pattern Recognition (ACPR) pp. 735-739

    4. Amrita Chaudhary, Swarnendu Bag, Mousumi Mandal, Sri Phani Krishna Karri, Ananya Barui, Monika Rajput, Provas Banerjee, Debdoot Sheet, Jyotirmoy Chatterjee, "Modulating prime molecular expressions and in vitro wound healing rate in keratinocyte (HaCaT) population under characteristic honey dilutions", Journal of ethnopharmacology pp. 211-219

    5. SL Happy, Swarnadip Chatterjee, Debdoot Sheet, "Unsupervised segmentation of overlapping cervical cell cytoplasm", arXiv preprint arXiv:1505.05601

    6. Abhijit Guha Roy, Sailesh Conjeti, Stephane G Carlier, Andreas König, Adnan Kastrati, Pranab K Dutta, Andrew F Laine, Nassir Navab, Debdoot Sheet, Amin Katouzian, "Bag of forests for modelling of tissue energy interaction in optical coherence tomography for atherosclerotic plaque susceptibility assessment", 2015 IEEE 12th International Symposium on Biomedical Imaging (ISBI) pp. 428-431

    7. Hrushikesh Tukaram Garud, Debdoot Sheet, Ajoy Kumar Ray, Manjunatha Mahadevappa, Jyotirmoy Chatterjee, "Adaptive weighted-local-difference order statistics filters", US Patent 9208545

    8. Sailesh Conjeti, Mehmet Yigitsoy, Debdoot Sheet, Jyotirmoy Chatterjee, Nassir Navab, Amin Katouzian, "Mutually coherent structural representation for image registration through joint manifold embedding and alignment", 2015 IEEE 12th International Symposium on Biomedical Imaging (ISBI) pp. 601-604

    9. Kausik Basak, Goutam Dey, Debdoot Sheet, Manjunatha Mahadevappa, Mahitosh Mandal, Pranab K Dutta, "Probabilistic graphical modeling of speckle statistics in laser speckle contrast imaging for noninvasive and label-free retinal angiography", 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) pp. 6244-6247

    10. Sailesh Conjeti, Mehmet Yigitsoy, Tingying Peng, Debdoot Sheet, Jyotirmoy Chatterjee, Christine Bayer, Nassir Navab, Amin Katouzian, "Deformable registration of immunofluorescence and histology using iterative cross-modal propagation", 2015 IEEE 12th International Symposium on Biomedical Imaging (ISBI) pp. 310-313

    11. Satarupa Banerjee, Debdoot Sheet, Amita Giri, Ranjan Ghosh, Ranjan Rashmi Paul, Mousumi Pal, Jyotirmoy Chatterjee, "optical Coherence Tomographic Attenuation Imaging Based Oral Precancer Diagnosis - Op239.", Head & Neck pp. E133

    12. Hrushikesh Tukaram Garud, Debdoot Sheet, Ajoy Kumar Ray, Manjunatha Mahadevappa, Jyotirmoy Chatterjee, "Adaptive Weighted-Local-Difference Order Statistics Filters", US Patent 9208545

    13. Shanbao Tong, Debdoot Sheet, Shoaib Bhuiyan, Martha Lucía Zequer Diaz, Andrew Taberne, "BME Trends Around the World: From Baby X to frugal technologies, here's what biomedical engineers are excited about in 2015.[From the Editors]", IEEE pulse pp. 4-6

  • 2014:

    1. Debdoot Sheet, Athanasios Karamalis, Abouzar Eslami, Peter Noël, Jyotirmoy Chatterjee, Ajoy K Ray, Andrew F Laine, Stephane G Carlier, Nassir Navab, Amin Katouzian, "Joint learning of ultrasonic backscattering statistical physics and signal confidence primal for characterizing atherosclerotic plaques using intravascular ultrasound", Medical image analysis pp.103-117

    2. Debdoot Sheet, Athanasios Karamalis, Abouzar Eslami, Peter Noël, Renu Virmani, Masataka Nakano, Jyotirmoy Chatterjee, Ajoy K Ray, Andrew F Laine, Stephane G Carlier, Nassir Navab, Amin Katouzian, "Hunting for necrosis in the shadows of intravascular ultrasound", Computerized Medical Imaging and Graphics pp. 104-112

    3. Debdoot Sheet, Satarupa Banerjee, Sri Phani Krishna Karri, Swarnendu Bag, Anji Anura, Amita Giri, Ranjan Rashmi Paul, Mousumi Pal, Badal C Sarkar, Ranjan Ghosh, Amin Katouzian, Nassir Navab, Ajoy K Ray, "Transfer learning of tissue photon interaction in optical coherence tomography towardsin vivo histology of the oral mucosa", 2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI) pp. 1389-1392

    4. Sri Phani Krishna Karri, Hrushikesh Garud, Debdoot Sheet, Jyotirmoy Chatterjee, Debjani Chakraborty, Ajoy Kumar Ray, Manjunatha Mahadevappa, "Learning scale-space representation of nucleus for accurate localization and segmentation of epithelial squamous nuclei in cervical smears", IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI) pp. 772-775

    5. Hrushikesh Garud, Debdoot Sheet, Amit Suveer, Manjunatha Mahadevappa, Ajoy Kumar Ray, "Method and apparatus for enhancing representations of micro-calcifications in a digital mammogram image", US Patent 8634630

    6. Hrushikesh Garud, Debdoot Sheet, Amit Suveer, Manjunatha Mahadevappa, Ajoy Kumar Ray, "Method and apparatus for enhancing representations of micro-calcifications in a digital mammogram image", US Patent