Dr. Imon Banerjee Profile

Dr. Imon Banerjee

Associate Professor at Mayo Clinic

SPIE Involvement:

Author

SPIE Journal Paper | 9 November 2024

Self-supervised learning for chest computed tomography: training strategies and effect on downstream applications

Amara Tariq, Gokul Ramasamy, Bhavik Patel, Imon Banerjee

JMI, Vol. 11, Issue 06, 064003, (November 2024) https://doi.org/10.1117/12.10.1117/1.JMI.11.6.064003

Read Abstract +

Proceedings Article | 3 April 2024 Poster + Paper

Comparative analysis of multiphase CT volumetric kidney segmentation: fine-tuning to domain adaptation

Ramon Correa-Medero, Sam Fathizadeh, Fatima Al Khafaji, Haidar Abdul-Muhsin, Bhavik Patel, Imon Banerjee

Proceedings Volume 12927, 129273A (2024) https://doi.org/10.1117/12.3003574

KEYWORDS: Kidney, Image segmentation, Computed tomography

Read Abstract +

Development of kidney segmentation models has largely focused on using contrast-enhanced CT exams. The KiTS segmentation challenge, in particular, has provided a benchmark using 300 annotated arterial phase CT scans. Review of the best performing models identifies 3D Unet models with residual connections as the best performing models for kidney segmentation. Li et al. found a U-net architecture with residual connections to provide the best performance for their segmentation task. Their work focused on segmenting kidney parenchyma alongside kidney stones using a dataset of 257 studies that was recently made available. Yu et al. investigated the ability to train a multi-organ nn-Unet model using simultaneous contrast and non-contrast images. The authors found the model to achieve high dice scores for kidney segmentation with average dice scores of 0.96.4 Inspection of output segmentation found the model to underperform for non-contrast images as measured by smaller quality assessment scores. Tang et al. took a similar approach where early and late arterial phase scans were used to train a patch-based network to segment renal structures. The group found the model to perform adequately on test data, with no distinction between late and early arterial phase performance. Lee et al. attempted to reduce the dependency of labeling multiple phases by using paired samples where only the contrast-enhanced volume was annotated. They were able to outperform existing models but were limited by the need for correct anatomical correspondence between scans. Ananda et al. removed the dependency for paired samples by training a dual discriminator-based network where the model is trained in three phases. One phase ensures the consistency of segmentation of contrast phase images; the following two phases ensure the image encoding and output maps are not significantly different from contrast and non-contrast images. Dinsdale et al. implemented a similar multi-step approach to improve segmentation quality by improving resiliency to agerelated physiological changes in Brain MRI, where cross entropy(CE) loss is utilized for training a discriminator on the patient’s age using features from the bottleneck and segmentation map. In contrast, the final phase uses a confusion loss, penalizing the model for having greater confidence for a particular age.8 Despite all the modeling effort, the application of evaluating patients with impaired kidney function is challenging since variation occurs not only due to phase of contrast but also the level of renal function, and thus the consistency of imaging appearance is not guaranteed. Establishing the need for techniques that would improve the robustness of kidney segmentation models. Within the scope of this work, we proposed various techniques to generate models resilient to different contrast phases and externally validated the models.

Proceedings Article | 3 April 2024 Presentation

Bias in radiology artificial intelligence: causes, evaluation and mitigation

Imon Banerjee

Proceedings Volume 12926, 129260O (2024) https://doi.org/10.1117/12.3023603

Read Abstract +

Proceedings Article | 2 April 2024 Poster + Paper

A study of relationship between social determinant of health and imaging based age estimation using head CT

Amara Tariq, Bhavik Patel, Imon Banerjee

Proceedings Volume 12926, 129262G (2024) https://doi.org/10.1117/12.3006496

KEYWORDS: Biological research, Biological imaging, Head, Computed tomography, Image processing, Education and training, X-ray computed tomography, Statistical analysis, Performance modeling, Neuroimaging

Read Abstract +

Proceedings Article | 2 April 2024 Presentation + Paper

Design, training, and applications of foundation model for chest computed tomography volumes

Amara Tariq, Bhavik Patel, Imon Banerjee

Proceedings Volume 12926, 1292611 (2024) https://doi.org/10.1117/12.3003042

KEYWORDS: Data modeling, Education and training, Lung, Chest, Visual process modeling, Computed tomography, Image segmentation, Performance modeling, Image processing

Read Abstract +

SPIE Journal Paper | 12 October 2023

Challenges and solutions of echocardiography generalization for deep learning: a study in patients with constrictive pericarditis

Jiwoong Jeong, Chieh-Ju Chao, Reza Arsanjani, Kihong Kim, Melissa Pelkey, Yi-Chieh Chen, Raheel Ramzan, Mohammad Elbahnasawy, Mohamed Sleem, Chadi Ayoub, Juan Maria Farina, Martha Grogan, Garvan Kane, Bhavik Patel, Jae Oh, Imon Banerjee

JMI, Vol. 10, Issue 05, 054502, (October 2023) https://doi.org/10.1117/12.10.1117/1.JMI.10.5.054502

KEYWORDS: Data modeling, Education and training, RGB color model, Echocardiography, Performance modeling, Deep learning, Motion models, Ablation, 3D modeling, Image classification

Read Abstract +

SPIE Journal Paper | 28 June 2023

Fusion of imaging and non-imaging data for disease trajectory prediction for coronavirus disease 2019 patients

Amara Tariq, Siyi Tang, Hifza Sakhi, Leo Anthony Celi, Janice Newsome, Daniel Rubin, Hari Trivedi, Judy Gichoya, Imon Banerjee

JMI, Vol. 10, Issue 03, 034004, (June 2023) https://doi.org/10.1117/12.10.1117/1.JMI.10.3.034004

KEYWORDS: Data modeling, Data fusion, Diseases and disorders, Image fusion, Chest imaging, Performance modeling, Modeling, Machine learning, COVID 19, Education and training

Read Abstract +

Proceedings Article | 10 April 2023 Presentation + Paper

A robust two-step adversarial debiasing with partial learning: medical image case-studies

Ramon Correa, Jiwoong Jason Jeong, Bhavik Patel, Hari Trivedi, Judy Gichoya, Imon Banerjee

Proceedings Volume 12469, 1246908 (2023) https://doi.org/10.1117/12.2647285

KEYWORDS: Artificial intelligence, Mammography, Breast density, Chest imaging

Read Abstract +

SPIE Journal Paper | 23 June 2022

GAN augmentation for multiclass image classification using hemorrhage detection as a case-study

Jiwoong Jeong, Bhavik Patel, Imon Banerjee

JMI, Vol. 9, Issue 03, 035504, (June 2022) https://doi.org/10.1117/12.10.1117/1.JMI.9.3.035504

KEYWORDS: Gallium nitride, Data modeling, Image classification, Computed tomography, Brain, Tissues, Binary data, Performance modeling, Neuroimaging, Medical imaging

Read Abstract +

SPIE Journal Paper | 3 February 2022

Margin-aware intraclass novelty identification for medical images

Xiaoyuan Guo, Judy Gichoya, Saptarshi Purkayastha, Imon Banerjee

JMI, Vol. 9, Issue 01, 014004, (February 2022) https://doi.org/10.1117/12.10.1117/1.JMI.9.1.014004

KEYWORDS: Data modeling, Medical imaging, Binary data, Performance modeling, Data centers, Visualization, Sensors, Visual process modeling, Tumor growth modeling, Distortion

Read Abstract +

Purpose: Existing anomaly detection methods focus on detecting interclass variations while medical image novelty identification is more challenging in the presence of intraclass variations. For example, a model trained with normal chest x-ray and common lung abnormalities is expected to discover and flag idiopathic pulmonary fibrosis, which is a rare lung disease and unseen during training. The nuances of intraclass variations and lack of relevant training data in medical image analysis pose great challenges for existing anomaly detection methods. Approach: We address the above challenges by proposing a hybrid model—transformation-based embedding learning for novelty detection (TEND), which combines the merits of classifier-based approach and AutoEncoder (AE)-based approach. Training TEND consists of two stages. In the first stage, we learn in-distribution embeddings with an AE via the unsupervised reconstruction. In the second stage, we learn a discriminative classifier to distinguish in-distribution data and the transformed counterparts. Additionally, we propose a margin-aware objective to pull in-distribution data in a hypersphere while pushing away the transformed data. Eventually, the weighted sum of class probability and the distance to margin constitutes the anomaly score. Results: Extensive experiments are performed on three public medical image datasets with the one-vs-rest setup (namely one class as in-distribution data and the left as intraclass out-of-distribution data) and the rest-vs-one setup. Additional experiments on generated intraclass out-of-distribution data with unused transformations are implemented on the datasets. The quantitative results show competitive performance as compared to the state-of-the-art approaches. Provided qualitative examples further demonstrate the effectiveness of TEND. Conclusion: Our anomaly detection model TEND can effectively identify the challenging intraclass out-of-distribution medical images in an unsupervised fashion. It can be applied to discover unseen medical image classes and serve as the abnormal data screening for downstream medical tasks. The corresponding code is available at https://github.com/XiaoyuanGuo/TEND_MedicalNoveltyDetection.

SPIE Journal Paper | 2 November 2017

Assessing treatment response in triple-negative breast cancer from quantitative image analysis in perfusion magnetic resonance imaging

Imon Banerjee, Sadhika Malladi, Daniela Lee, Adrien Depeursinge, Melinda Telli, Jafi Lipson, Daniel Golden, Daniel Rubin

JMI, Vol. 5, Issue 01, 011008, (November 2017) https://doi.org/10.1117/12.10.1117/1.JMI.5.1.011008

KEYWORDS: Breast cancer, Performance modeling, Tumors, Magnetic resonance imaging, Data modeling, Wavelets, Statistical modeling, Image filtering, Statistical analysis, Breast

Read Abstract +

Showing 5 of 11 publications

Conference Committee Involvement (1)

Image Processing

17 February 2025 | San Diego, California, United States

Keywords/Phrases

Search In:

Publication Years