A5008932408- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Advanced Neural Network Applications
- Glioma Diagnosis and Treatment
- Cutaneous Melanoma Detection and Management
- Nonmelanoma Skin Cancer Studies
- Smart Agriculture and AI
- Medical Imaging and Analysis
- Video Surveillance and Tracking Methods
- Retinal Imaging and Analysis
- Brain Tumor Detection and Classification
- Autonomous Vehicle Technology and Safety
- Cell Image Analysis Techniques
- Advanced Image and Video Retrieval Techniques
- Renal cell carcinoma treatment
- Remote Sensing and LiDAR Applications
- Wireless Body Area Networks
- Glaucoma and retinal disorders
- Retinal Diseases and Treatments
- Remote Sensing in Agriculture
- Colorectal Cancer Screening and Detection
- Retinal and Optic Conditions
- Modular Robots and Swarm Intelligence
- Traffic Prediction and Management Techniques
- Bluetooth and Wireless Communication Technologies
Indiana University School of Medicine
2024-2025
Indiana University – Purdue University Indianapolis
2024-2025
University of Indianapolis
2025
Indiana University Indianapolis
2025
University of Pennsylvania
2023-2024
Sanskriti Samvardhan Mandal
2020-2022
Since the last few decades, number of road causalities has seen continuous growth across globe. Nowa-days intelligent transportation systems are being developed to enable safe and relaxed driving scene understanding surrounding environment is an integral part it. While several approaches for semantic segmentation based on deep learning Convolutional Neural Network (CNN), these assume well structured infrastructure environment. We focus our work recent India Driving Lite Dataset (IDD), which...
Age-related macular degeneration (AMD) is the leading cause of visual impairment among elderly in world. Early detection AMD great importance, as vision loss caused by this disease irreversible and permanent. Color fundus photography most cost-effective imaging modality to screen for retinal disorders. Cutting edge deep learning based algorithms have been recently developed automatically detecting from images. However, there are still lack a comprehensive annotated dataset standard...
Abstract Skin cancer is a serious condition that requires accurate diagnosis and treatment. One way to assist clinicians in this task using computer-aided tools automatically segment skin lesions from dermoscopic images. We propose novel adversarial learning-based framework called Efficient-GAN (EGAN) uses an unsupervised generative network generate lesion masks. It consists of generator module with top-down squeeze excitation-based compound scaled path, asymmetric lateral connection-based...
Chronic kidney disease (CKD) is a major global health issue, affecting over 10% of the population and causing significant mortality. While biopsy remains gold standard for CKD diagnosis treatment, lack comprehensive benchmarks pathology segmentation hinders progress in field. To address this, we organized Kidney Pathology Image Segmentation (KPIs) Challenge, introducing dataset that incorporates preclinical rodent models with 10,000 annotated glomeruli from 60+ Periodic Acid Schiff...
Distinguishing between renal oncocytic tumors, such as oncocytoma (RO), and a subset of tumors with overlapping characteristics, including the recently identified low-grade tumor (LOT), can present diagnostic challenge for pathologists owing to shared histopathologic features. To develop an automatic computational classifier stratifying whole slide images biopsy resection specimens into 2 distinct groups: RO LOT. A total 269 from 125 cases across 6 institutions were collected. weakly...
Abstract Diagnosis of diffuse glioma according to the WHO 2021 classification criteria mandate integration histologic features with molecular profiling. However, profiling is expensive, time-demanding, and when not available leads ‘not-otherwise-specified' status. We seek interpretable AI-based glioma, as oligodendroglioma, astrocytoma, or glioblastoma, from H&E-stained slides alone. identified 2, 114 multi-institutional whole slide images (WSIs), two independent retrospective...
The first Agriculture-Vision Challenge aims to encourage research in developing novel and effective algorithms for agricultural pattern recognition from aerial images, especially the semantic segmentation task associated with our challenge dataset. Around 57 participating teams various countries compete achieve state-of-the-art agriculture segmentation. Agriculture- Vision Dataset was employed, which comprises of 21,061 multi-spectral farmland images. This paper provides a summary notable...
With the rapid development of artificial intelligence (AI) in medical image processing, deep learning color fundus photography (CFP) analysis is also evolving. Although there are some open-source, labeled datasets CFPs ophthalmology community, large-scale for screening only have labels disease categories, and with annotations structures usually small size. In addition, labeling standards not uniform across datasets, no clear information on acquisition device. Here we release a...
Deep learning and pattern recognition in smart farming has seen rapid growth as a building bridge between crop science computer vision. One of the important application is anomaly segmentation agriculture like weed, standing water, cloud shadow, etc. Our research work focuses on aerial farmland image dataset known Agriculture Vision. We propose to have data fusion R, G, B, NIR modalities that enhances feature extraction also Efficient Fused Pyramid Network (Fuse-PN) for segmentation. The...
Introduction Glioblastoma (GBM) is a highly aggressive malignant tumor of the central nervous system that displays varying molecular and morphological profiles, leading to challenging prognostic assessments. Stratifying GBM patients according overall survival (OS) from H&E-stained whole slide images (WSI) using advanced computational methods challenging, but with direct clinical implications. Methods This work focusing on (IDH-wildtype, CNS WHO Gr.4) cases, identified TCGA-GBM...
Abstract BACKGROUND Glioblastoma is the most common malignant adult brain tumor with a poor prognosis and heterogeneous morphology. Stratifying glioblastoma patients according to overall survival (OS) from H&E-stained histopathology whole slide images (WSI) using advanced computational methods challenging task direct clinical implications. We hypothesize that quantifying morphology patterns present in WSI can yield biomarkers of prognostic relevance contributing optimizing...
Abstract BACKGROUND Glioblastoma is the most common malignant adult brain tumor, with a grim prognosis and heterogeneous morphology. Stagnant patient prospects over last 20 years reflect our limited disease understanding. Robust prognostic stratification from whole slide images (WSI) using interpretable computational methods could improve understanding management. MATERIAL AND METHODS Assessment of TCGA-GBM TCGA-LGG data collections by 2021 WHO classification criteria CNS tumors identified...
Abstract BACKGROUND Glioblastoma is the most common adult primary brain tumor. Limited improvement in patient prospects over last 20 years reflects our limited disease understanding. Robust prognostic stratification from whole slide images (WSI) & clinical data using interpretable computational methods could improve understanding and management. MATERIAL AND METHODS The TCGA-GBM TCGA-LGG collections were reclassified according to 2021 WHO classification criteria, identifying 188...
Glioblastoma is the most common and aggressive malignant adult tumor of central nervous system, with a grim prognosis heterogeneous morphologic molecular profiles. Since adopting current standard-of-care treatment 18 years ago, no substantial prognostic improvement has been noticed. Accurate prediction patient overall survival (OS) from histopathology whole slide images (WSI) integrated clinical data using advanced computational methods could optimize decision-making management. Here, we...
Abstract BACKGROUND Infiltrating gliomas are the most common primary adult brain tumors. Isocitrate dehydrogenase (IDH) mutation is a key driver of gliomagenesis in 25–30% infiltrating and correlates with favorable prognosis when compared histologically-similar but biologically-distinct IDH-wildtype glioblastoma. We sought an interpretable computational pipeline to predict IDH status from digitized histopathology glioma sections (Whole Slide Images - WSI) identify associated cell-level...
Glioblastoma is the most common primary adult brain tumor, with a grim prognosis - median survival of 12-18 months following treatment, and 4 otherwise. widely infiltrative in cerebral hemispheres well-defined by heterogeneous molecular micro-environmental histopathologic profiles, which pose major obstacle treatment. Correctly diagnosing these tumors assessing their heterogeneity crucial for choosing precise treatment potentially enhancing patient rates. In gold-standard...
Abstract Glioblastoma, the most common malignant primary adult brain tumor, poses significant diagnostic and treatment challenges due to its heterogeneous molecular micro-environmental profiles. To this end, we organize BraTS-Path challenge provide a public benchmarking environment comprehensive dataset develop validate AI models for identifying distinct histopathologic glioblastoma sub-regions in H&E-stained digitized tissue sections. We identified 188 multi-institutional slides of...
Uncertainty in medical image segmentation tasks, especially inter-rater variability, arising from differences interpretations and annotations by various experts, presents a significant challenge achieving consistent reliable segmentation. This variability not only reflects the inherent complexity subjective nature of interpretation but also directly impacts development evaluation automated algorithms. Accurately modeling quantifying this is essential for enhancing robustness clinical...
Abstract INTRODUCTION Isocitrate-dehydrogenase (IDH) mutational status is diagnostically critical in adult gliomas, with prognostic and therapeutic implications. IDH cannot be determined by sole histologic assessment requires molecular testing. We hypothesize that AI-based analysis can accurately predict glioma, retrieving sub-visual cues H&E-stained digitized tissue sections. MATERIALS AND METHODS A retrospective discovery cohort of 1,534 cases (756/778 IDH-wildtype/IDH-mutant,...
The first Agriculture-Vision Challenge aims to encourage research in developing novel and effective algorithms for agricultural pattern recognition from aerial images, especially the semantic segmentation task associated with our challenge dataset. Around 57 participating teams various countries compete achieve state-of-the-art agriculture segmentation. Dataset was employed, which comprises of 21,061 multi-spectral farmland images. This paper provides a summary notable methods results...