Posts by Collection

Benchmarking Encoder-Decoder Architectures for Biplanar X-ray to 3D Shape Reconstruction

Published in NeurIPS 2023 Datasets and Benchmark Track, 2023

We provide benchmarking toolkit, including architecture implementations, clinical metric evaluation, benchmarking tasks, data preprocessing utils for Biplanar X-ray to 3D Shape Reconstruction.

Recommended citation: Shakya, Mahesh, and Bishesh Khanal. "Benchmarking Encoder-Decoder Architectures for Biplanar X-ray to 3D Bone Shape Reconstruction." (2023) https://arxiv.org/pdf/2309.13587

From Development to Deployment of AI-Assisted Telehealth and Screening for Vision and Hearing-Threatening Diseases in Resource-Constrained Settings: Field Observations, Challenges and Way Forward

Published in International Workshop on Medical Image Computing in Resource Constrained Settings (MIRASOL), Springer, 2026, 2026

Vision- and hearing-threatening diseases cause preventable disability, especially where specialists and screening facilities are scarce. AI-assisted screening and telehealth could widen early detection, but deploying them in paper-based workflows is hard and little field experience has been documented. This paper reports field observations from developing and deploying such a system, the challenges met, and the way forward.

Recommended citation: Shakya, Mahesh, Bijay Adhikari, Nirsara Shrestha, Bipin Koirala, Arun Adhikari, Prasanta Poudyal, Luna Mathema, Sarbagya Buddhacharya, Bijay Khatri, and Bishesh Khanal. "From Development to Deployment of AI-Assisted Telehealth and Screening for Vision and Hearing-Threatening Diseases in Resource-Constrained Settings: Field Observations, Challenges and Way Forward." (2026) https://link.springer.com/chapter/10.1007/978-3-032-13654-1_31

Large-Scale Pretraining for Improving Deep Learning-Based Geometric Distortion Correction of Diffusion-Weighted Imaging

Published in arXiv preprint arXiv:2609.06437, 2026

We study whether large-scale pretraining (a self-supervised model and a generative model) improves deep learning-based correction of geometric distortion in single-phase-encoding diffusion-weighted MRI, compared with a non-pretrained baseline. Pretrained models do better, but applying the best one to data from a low-resource setting shows transferability problems, which registration to a common standard space helps to reduce.

Recommended citation: Khanal, Saroj, Yashawant Kumar Yadav, Kritam Bhattarai, Jeevan Neupane, Shristi Subedi, Saship Gwachha, Manish Kumar Tiwari, Dong Zhang, Confidence Raymond, Aondona Moses Iorumbur, Udunna Anazodo, Surendra Maharjan, Bishesh Khanal, Mahesh Shakya, and Pralhad Kumar Shrestha. "Large-Scale Pretraining for Improving Deep Learning-Based Geometric Distortion Correction of Diffusion-Weighted Imaging." arXiv:2609.06437 (2026) https://arxiv.org/abs/2609.06437

AI and Medical Image Informatics

Published: July 19, 2023

Showcased our (then) current work in Biplanar X-ray to 3D Bone Shape Reconstruction. Explained what Medical Image Informatics is and how AI and specifically Deep Learning (Learning from Data) has improved the state-of-the-art in Vision tasks. Additionally, described the challenges in bringing these breakthroughs into the clinician’s hands including robustness, uncertainty estimation, out-of-distribution(OOD) issues.

Medical Image Analysis: 2D to 3D Reconstruction

Published: August 14, 2023

Showcased our (then) current work in Biplanar X-ray to 3D Bone Shape Reconstruction.

OpenScience(OpenCode+OpenData+OpenPaper)

Published: October 12, 2023

Described what we do at NAAMII and how we are committed towards OpenScience. This included my own experience with facing reproducibility issues when exploring relevant literature in Biplanar X-ray to 3D Shape Reconstruction and how we have tried to address that by providing an opensource framework to the community for other researchers to build upon.