no code implementations • 22 Mar 2024 • Shreyas Chandgothia, Ardhendu Sekhar, Amit Sethi
Few-shot segmentation techniques reduce the required number of images to learn to segment a new class, but careful annotations of object boundaries are still required.
no code implementations • 4 Mar 2024 • Akhila Krishna, Ravi Kant Gupta, Pranav Jeevan, Amit Sethi
Gene selection plays a pivotal role in oncology research for improving outcome prediction accuracy and facilitating cost-effective genomic profiling for cancer patients.
no code implementations • 19 Feb 2024 • Pirzada Suhail, Supratik Chakraborty, Amit Sethi
While the deployment of neural networks, yielding impressive results, becomes more prevalent in various applications, their interpretability and understanding remain a critical challenge.
no code implementations • 30 Nov 2023 • Sahar Almahfouz Nasser, Shashwat Pathak, Keshav Singhal, Mohit Meena, Nihar Gupte, Ananya Chinmaya, Prateek Garg, Amit Sethi
One approach involves converting images into nodes by identifying significant keypoints within them.
1 code implementation • 7 Nov 2023 • Shruti Sawant, Jeslin Thalapil, Siddharth Tallur, Sauvik Banerjee, Amit Sethi
We believe that we are the first to report numerical models for four types of damages in HCSS, which is followed up with experimental validation.
no code implementations • 5 Oct 2023 • Amruta Parulekar, Utkarsh Kanwat, Ravi Kant Gupta, Medha Chippa, Thomas Jacob, Tripti Bameta, Swapnil Rane, Amit Sethi
We propose a method to train DNNs for instance segmentation and classification on multiple datasets where the set of classes across the datasets are related but not the same.
no code implementations • 29 Sep 2023 • Ravi Kant Gupta, Shounak Das, Amit Sethi
This paper presents a novel approach for unsupervised domain adaptation (UDA) targeting H&E stained histology images.
no code implementations • 10 Aug 2023 • Sahar Almahfouz Nasser, Ashutosh Sharma, Anmol Saraf, Amruta Mahendra Parulekar, Purvi Haria, Amit Sethi
Then, we utilize generative adversarial networks (GANs) to tackle the inverse problem of generating mammogram-quality images from ultrasound images.
1 code implementation • 20 Jul 2023 • Sahar Almahfouz Nasser, Nihar Gupte, Amit Sethi
We propose a novel approach based on reverse knowledge distillation to train large models with limited data while preventing overfitting.
Ranked #1 on Image Registration on FIRE
no code implementations • 16 Jul 2023 • Akhila Krishna K, Ravi Kant Gupta, Nikhil Cherian Kurian, Pranav Jeevan, Amit Sethi
The heterogeneity of breast cancer presents considerable challenges for its early detection, prognosis, and treatment selection.
1 code implementation • 1 Jul 2023 • Pranav Jeevan, Akella Srinidhi, Pasunuri Prathiba, Amit Sethi
We propose a new neural network -- WaveMixSR -- for image super-resolution based on WaveMix architecture which uses a 2D-discrete wavelet transform for spatial token-mixing.
Ranked #1 on Image Super-Resolution on BSD100 - 2x upscaling
1 code implementation • 1 Jul 2023 • Pranav Jeevan, Dharshan Sampath Kumar, Amit Sethi
The current state-of-the-art models for image inpainting are computationally heavy as they are based on transformer or CNN backbones that are trained in adversarial or diffusion settings.
Ranked #1 on Image Inpainting on ImageNet
1 code implementation • 29 May 2023 • Philippe Weitz, Masi Valkonen, Leslie Solorzano, Circe Carr, Kimmo Kartasalo, Constance Boissin, Sonja Koivukoski, Aino Kuusela, Dusan Rasic, Yanbo Feng, Sandra Sinius Pouplier, Abhinav Sharma, Kajsa Ledesma Eriksson, Stephanie Robertson, Christian Marzahl, Chandler D. Gatenbee, Alexander R. A. Anderson, Marek Wodzinski, Artur Jurgas, Niccolò Marini, Manfredo Atzori, Henning Müller, Daniel Budelmann, Nick Weiss, Stefan Heldmann, Johannes Lotz, Jelmer M. Wolterink, Bruno De Santi, Abhijeet Patil, Amit Sethi, Satoshi Kondo, Satoshi Kasai, Kousuke Hirasawa, Mahtab Farrokh, Neeraj Kumar, Russell Greiner, Leena Latonen, Anne-Vibeke Laenkholm, Johan Hartman, Pekka Ruusuvuori, Mattias Rantalainen
The alignment of tissue between histopathological whole-slide-images (WSI) is crucial for research and clinical applications.
no code implementations • 19 Apr 2023 • Chirag P, Mukta Wagle, Ravi Kant Gupta, Pranav Jeevan, Amit Sethi
We propose a new technique called CHATTY: Coupled Holistic Adversarial Transport Terms with Yield for Unsupervised Domain Adaptation.
Ranked #1 on Unsupervised Domain Adaptation on FHIST
no code implementations • 17 Mar 2023 • Nikhil Cherian Kurian, Varsha S, Abhijit PATIL, Shashikant Khade, Amit Sethi
The outlier score derived from the OOD detector is used to modulate sample selection for the subsequent semi-SL stage, ensuring that samples conforming to the distribution of the few labeled samples are more frequently exposed to the subsequent semi-SL framework.
no code implementations • 22 Feb 2023 • Pranav Jeevan, Nikhil Cherian Kurian, Amit Sethi
Convolution Neural Networks (CNNs) are widely used in medical image analysis, but their performance degrade when the magnification of testing images differ from the training images.
Ranked #1 on Image Classification on BreakHis
Breast Cancer Histology Image Classification Image Classification
no code implementations • 25 Nov 2022 • Harsh Shah, Thomas Jacob, Amruta Parulekar, Anjali Amarapurkar, Amit Sethi
In this study, we trained and tested a deep neural network based on UNet to detect and count eosinophils in GI tract biopsies.
1 code implementation • 15 Sep 2022 • Tirupati Saketh Chandr, Sahar Almahfouz Nasser, Nikhil Cherian Kurian, Amit Sethi
The effective localization of mitosis is a critical precursory task for deciding tumor prognosis and grade.
no code implementations • 26 Aug 2022 • Ravi Kant Gupta, Shivani Nandgaonkar, Nikhil Cherian Kurian, Swapnil Rane, Amit Sethi
With our pipeline, we achieved an average area under the curve (AUC) of 0. 964 for tumor detection, and 0. 942 for histological subtyping between adenocarcinoma and squamous cell carcinoma on the TCGA dataset.
no code implementations • 18 Jul 2022 • Bibin Wilson, Anand Singh, Amit Sethi
We discuss the adaptation of equipment and sensors for the collection of navigation, control, and bathymetry data and also give an overview of the vehicle setup.
no code implementations • 5 Jul 2022 • Bibin Wilson, Rajiv Kumar, Narayanarao Bhogapurapu, Anand Singh, Amit Sethi
DI-UNet uses a deep learning architecture to predict the resistivity using full polarimetric SAR data by promising a quick survey addition to the traditional method.
1 code implementation • 28 May 2022 • Pranav Jeevan, Kavitha Viswanathan, Anandu A S, Amit Sethi
The whole architecture is a stack of self-similar and resolution-preserving WaveMix blocks, which allows architectural flexibility for various tasks and levels of resource availability.
Ranked #1 on Image Classification on Galaxy10 DECals (using extra training data)
no code implementations • 3 May 2022 • Nikhil Cherian Kurian, Amit Lohan, Gregory Verghese, Nimish Dharamshi, Swati Meena, Mengyuan Li, Fangfang Liu, Cheryl Gillet, Swapnil Rane, Anita Grigoriadis, Amit Sethi
Although the U-Net architecture has been extensively used for segmentation of medical images, we address two of its shortcomings in this work.
1 code implementation • 7 Mar 2022 • Pranav Jeevan, Amit Sethi
The multi-scale nature of the DWT also reduces the requirement for a deeper architecture compared to the CNNs, as the latter relies on pooling for partial spatial mixing.
Ranked #15 on Image Classification on MNIST
no code implementations • 5 Mar 2022 • Sahar Almahfouz Nasser, Nikhil Cherian Kurian, Saqib Shamsi, Mohit Meena, Amit Sethi
We present WSSAMNet, a weakly supervised method for medical image registration.
1 code implementation • 25 Jan 2022 • Pranav Jeevan, Amit Sethi
Vision transformers (ViTs) have found only limited practical use in processing images, in spite of their state-of-the-art accuracy on certain benchmarks.
Ranked #22 on Image Classification on Tiny ImageNet Classification
1 code implementation • 23 Jan 2022 • Sahar Almahfouz Nasser, Saqib Shamsi, Valay Bundele, Bhavesh Garg, Amit Sethi
Capturing high-resolution magnetic resonance (MR) images is a time consuming process, which makes it unsuitable for medical emergencies and pediatric patients.
no code implementations • 13 Dec 2021 • Bhakti Baheti, Satrajit Chakrabarty, Hamed Akbari, Michel Bilello, Benedikt Wiestler, Julian Schwarting, Evan Calabrese, Jeffrey Rudie, Syed Abidi, Mina Mousa, Javier Villanueva-Meyer, Brandon K. K. Fields, Florian Kofler, Russell Takeshi Shinohara, Juan Eugenio Iglesias, Tony C. W. Mok, Albert C. S. Chung, Marek Wodzinski, Artur Jurgas, Niccolo Marini, Manfredo Atzori, Henning Muller, Christoph Grobroehmer, Hanna Siebert, Lasse Hansen, Mattias P. Heinrich, Luca Canalini, Jan Klein, Annika Gerken, Stefan Heldmann, Alessa Hering, Horst K. Hahn, Mingyuan Meng, Lei Bi, Dagan Feng, Jinman Kim, Ramy A. Zeineldin, Mohamed E. Karar, Franziska Mathis-Ullrich, Oliver Burgert, Javid Abderezaei, Aymeric Pionteck, Agamdeep Chopra, Mehmet Kurt, Kewei Yan, Yonghong Yan, Zhe Tang, Jianqiang Ma, Sahar Almahfouz Nasser, Nikhil Cherian Kurian, Mohit Meena, Saqib Shamsi, Amit Sethi, Nicholas J. Tustison, Brian B. Avants, Philip Cook, James C. Gee, Lin Tian, Hastings Greer, Marc Niethammer, Andrew Hoopes, Malte Hoffmann, Adrian V. Dalca, Stergios Christodoulidis, Theo Estiene, Maria Vakalopoulou, Nikos Paragios, Daniel S. Marcus, Christos Davatzikos, Aristeidis Sotiras, Bjoern Menze, Spyridon Bakas, Diana Waldmannstetter
Registration of longitudinal brain MRI scans containing pathologies is challenging due to dramatic changes in tissue appearance.
no code implementations • 21 Oct 2021 • Himanshu Pradeep Aswani, Abhiraj Sunil Kanse, Shubhang Bhatnagar, Amit Sethi
Training CNNs from scratch on new domains typically demands large numbers of labeled images and computations, which is not suitable for low-power hardware.
1 code implementation • 29 Sep 2021 • Pranav Jeevan P, Amit Sethi
Our work suggests that research on model structures that exploit the right inductive bias is far from over, and that such models can enable the training of computer vision models in settings with limited GPU resources.
Ranked #203 on Image Classification on CIFAR-10
2 code implementations • 5 Jul 2021 • Pranav Jeevan, Amit Sethi
Secondly, we introduced an inductive bias for images by replacing the initial linear embedding layer by convolutional layers in ViX, which significantly increased classification accuracy without increasing the model size.
Ranked #206 on Image Classification on CIFAR-10 (using extra training data)
no code implementations • 30 Nov 2020 • Abhijeet Patil, Mohd. Talha, Aniket Bhatia, Nikhil Cherian Kurian, Sammed Mangale, Sunil Patel, Amit Sethi
Performance of deep learning algorithms decreases drastically if the data distributions of the training and testing sets are different.
no code implementations • 29 Oct 2020 • Shubhang Bhatnagar, Sachin Goyal, Darshan Tank, Amit Sethi
To counter the paucity of data, we also deploy another head on the scoring network for regularization via multi-task learning and use an unusual self-balancing hybrid scoring function.
1 code implementation • 16 Sep 2020 • Himanshu Pradeep Aswani, Amit Sethi
Current research suggests that the key factors in designing neural network architectures involve choosing number of filters for every convolution layer, number of hidden neurons for every fully connected layer, dropout and pruning.
no code implementations • 9 Aug 2020 • Deepak Anand, Gaurav Patel, Yaman Dang, Amit Sethi
Remarkably, without retraining on target datasets, our pre-trained nucleus detector also outperformed existing nucleus detectors that were trained on at least some of the images from the target datasets.
no code implementations • 16 Jun 2020 • Mookund Sureka, Abhijeet Patil, Deepak Anand, Amit Sethi
With the increase in the use of deep learning for computer-aided diagnosis in medical images, the criticism of the black-box nature of the deep learning models is also on the rise.
no code implementations • MIDL 2019 • Pranav Poduval, Hrushikesh Loya, Amit Sethi
Deep neural networks have revolutionized medical image analysis and disease diagnosis.
no code implementations • 6 Apr 2020 • Mukesh Kumar Vishal, Dipesh Tamboli, Abhijeet Patil, Rohit Saluja, Biplab Banerjee, Amit Sethi, Dhandapani Raju, Sudhir Kumar, R N Sahoo, Viswanathan Chinnusamy, J Adinarayana
The present investigation is carried out for discriminating drought tolerant, and susceptible genotypes.
no code implementations • 19 Mar 2020 • Hrushikesh Loya, Pranav Poduval, Deepak Anand, Neeraj Kumar, Amit Sethi
Survival models are used in various fields, such as the development of cancer treatment protocols.
1 code implementation • 16 Feb 2020 • Abhijeet Patil, Dipesh Tamboli, Swati Meena, Deepak Anand, Amit Sethi
We aim to provide a better interpretation of classification results by providing localization on microscopic histopathology images.
no code implementations • 21 Aug 2019 • Yaman Dang, Deepak Anand, Amit Sethi
One of the first steps in the diagnosis of most cardiac diseases, such as pulmonary hypertension, coronary heart disease is the segmentation of ventricles from cardiac magnetic resonance (MRI) images.
no code implementations • 14 Aug 2019 • Shrey Gadiya, Deepak Anand, Amit Sethi
Spatial arrangement of cells of various types, such as tumor infiltrating lymphocytes and the advancing edge of a tumor, are important features for detecting and characterizing cancers.
1 code implementation • 10 Jan 2019 • Goutham Ramakrishnan, Deepak Anand, Amit Sethi
Normalizing unwanted color variations due to differences in staining processes and scanner responses has been shown to aid machine learning in computational pathology.
no code implementations • 31 Oct 2018 • Shrey Gadiya, Deepak Anand, Amit Sethi
While convolutional neural networks (CNNs) have recently made great strides in supervised classification of data structured on a grid (e. g. images composed of pixel grids), in several interesting datasets, the relations between features can be better represented as a general graph instead of a regular grid.
1 code implementation • 22 Feb 2018 • Aditya Golatkar, Deepak Anand, Amit Sethi
In this paper, we propose a deep learning-based method for classification of H&E stained breast tissue images released for BACH challenge 2018 by fine-tuning Inception-v3 convolutional neural network (CNN) proposed by Szegedy et al.
1 code implementation • IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops 2017 • Kothapalli Vignesh, Gaurav Yadav, Amit Sethi
Next, features are extracted from each frame using a convolutional neural network (CNN) that is trained to classify between normal and abnormal frames.
Abnormal Event Detection In Video Anomaly Detection In Surveillance Videos +1
no code implementations • 13 Dec 2015 • Neeraj Kumar, Animesh Karmakar, Ranti Dev Sharma, Abhinav Mittal, Amit Sethi
We propose a method to classify images from target classes with a small number of training examples based on transfer learning from non-target classes.