member portrait Meghal Dani PhD Student International Max-Planck Research School for Intelligent Systems (IMPRS-IS)

Profile

Meghal Dani is a Ph.D. candidate under the Else Kröner-Fresenius Foundation (EKFS) grant at the International Max-Planck Research School for Intelligent Systems (IMPRS-IS). She is associated with EML and is supervised by Prof. Dr. Zeynep Akata and Dr. rer. nat. Stefanie Liebe respectively (*Earlier also worked at MIDAS lab with Prof. Dr. Med. Sergios Gatidis, who moved to Stanford in 2023).

She received her bachelor’s degree in Computer Science Engineering from BIT Mesra in 2016 and focused on AI in healthcare during her master’s at IIIT Delhi, majoring in Computational Biology (2017-2019). Since then, she has been interested in developing modern deep learning systems in healthcare that have the potential to make diagnostic decisions of similar quality to attending doctors. She worked as a full-time researcher (2019-2021) at Tata Research and Innovation Labs in the area of Deep Learning and AI under the supervision of Senior Scientist Dr. Lovekesh Vig and Ramya Hebbalaguppe.

Her primary research interests are addressing issues of transparency, robustness, interpretability, and reliability by combining vision and language models.

Publications

SemioLLM: Assessing Large Language Models for Semiological Analysis in Epilepsy Research
Meghal Dani, Muthu Jeyanthi Prakash, Zeynep Akata, and Stefanie Liebe
ICML AI4Science Workshop
[Paper](https://arxiv.org/pdf/2407.03004) | Code coming up soon

 

DeViL: Decoding Vision features into Language
Meghal Dani, Isabel Rio-Torto, Stephan Alaniz, and Zeynep Akata
German Conference on Pattern Recognition, 2023 (Oral)
[Paper](https://www.dagm-gcpr.de/fileadmin/dagm-gcpr/pictures/2023_Heidelberg/Paper_FastTrack/094.pdf) | [Code](https://github.com/ExplainableML/DeViL) | This was also presented at the ICCV 2023 Workshop on Closing the Loop Between Vision and Language (CLVL).

 

An Efficient Anchor-Free Universal Lesion Detection in CT-Scans
Manu Sheoran, Meghal Dani, Monika Sharma, Lovekesh Vig
International Symposium on Biomedical Imaging (ISBI), 2022
Paper

 

DKMA-ULD: Domain Knowledge augmented Multi-head Attention based Robust Universal Lesion Detection
Manu Sheoran, *Meghal Dani**, Monika Sharma, Lovekesh Vig
The British Machine Vision Conference (BMVC), 2021
Paper

 

3DPoseLite: A Compact 3D Pose Estimation Using Node Embeddings
Meghal Dani, Karan Narain, Ramya Hebbalaguppe
Winter Conference on Applications of Computer Vision (WACV), 2021
Paper

 

PoseFromGraph: Compact 3-D Pose Estimation using Graphs
Meghal Dani, Additya Popli, Ramya Hebbalaguppe
SIGGRAPH Asia, 2020
Paper

 

Mid-air fingertip-based user interaction in mixed reality
Meghal Dani, Gaurav Garg, Ramakrishna Perla, Ramya Hebbalaguppe
ISMAR, 2018
Paper Link

 

For up-to-date information and patents, please also check: Google Scholar.

Community Service

Reviewing

Other