Sdamirsa

Exminer

creating MCQs from your PDFs by leveraging LLMs.

AI-in-Med-Trend

Trend of AI in Medicine and GI

Exminer

creating MCQs from your PDFs by leveraging LLMs.

extract_telegram_contact_to_csv

This repo is created for other people looking to convert their telegram exported contacts to CSV (so they can import in in google account or elsewhere)

HappyTreeResearchFriends

Here I will store some learning contents, learning pathways, and opportunities so that I can re-use them. Happy Tree Research Friends is a group that I started three years ago with my mentee and friends.

IBD-ICD

IBD-ICD (Inflammatory Bowel Disease Imaging and Clinical Data) is a curated repository containing medical images and associated clinical datasets for research and analysis in the field of IBD.

PanCanAID

Pancreas Cancer AI-Driven Diagnosis in CT scan and EUS Imaging: A protocol for an Observational Multicentric Ambispective Study

Parsbench-Med-Eng

MedQA-but-better-yield (MedQA-BBY) is build on top of MedQA dataset but enriched with labels, as well as two correct options and one negative point option, and three 0 point options. The labels include:(1) question topic regarding system, discipline, and subspeciality, (2) token length, (3) Bloom taxonomy

Sdamirsa

This is me

Tehran_COVID_Cohort

Code for data extraction and analysis of a COVID project with 10,000 participants (COVID-19 inpatients)

Sdamirsa/HappyTreeResearchFriends

Here I will store some learning contents, learning pathways, and opportunities so that I can re-use them. Happy Tree Research Friends is a group that I started three years ago with my mentee and friends.

Sdamirsa/Parsbench-Med-Eng

MedQA-but-better-yield (MedQA-BBY) is build on top of MedQA dataset but enriched with labels, as well as two correct options and one negative point option, and three 0 point options. The labels include:(1) question topic regarding system, discipline, and subspeciality, (2) token length, (3) Bloom taxonomy

Sdamirsa/extract_telegram_contact_to_csv

This repo is created for other people looking to convert their telegram exported contacts to CSV (so they can import in in google account or elsewhere)

Sdamirsa/AI-in-Med-Trend

Trend of AI in Medicine and GI

Sdamirsa/Exminer

creating MCQs from your PDFs by leveraging LLMs.

Sdamirsa/ParsBench-biomedical

This repository contains multiple-choice questions and real-world scenarios for benchmarking large language models in Persian (Farsi) and English. The content is designed to evaluate the performance and capabilities of various language models across these two languages.

Sdamirsa/IBD-ICD

IBD-ICD (Inflammatory Bowel Disease Imaging and Clinical Data) is a curated repository containing medical images and associated clinical datasets for research and analysis in the field of IBD.

Sdamirsa/LLM-as-a-judge

LLM as a judge (developed by Nariman Naderi)

Sdamirsa/PanCanAID

Pancreas Cancer AI-Driven Diagnosis in CT scan and EUS Imaging: A protocol for an Observational Multicentric Ambispective Study

Sdamirsa/Tehran_COVID_Cohort

Code for data extraction and analysis of a COVID project with 10,000 participants (COVID-19 inpatients)

Sdamirsa/AI-and-Abdominal-CT-resources

Resources for people working on Artificial Intelligence (AI) and Machine Learning (ML) projects in computed tomography (CT) scan.

Sdamirsa/ehran_covid_Large-Language-Models-vs-Classical-Machine-learning

Sdamirsa/EXCT_v2

Extraction from clinical text (EXCT): update includes automated evaluation, parameter (prompt & temp) tuning, Pydantic structure slection, and Few-shot injection.

Sdamirsa/ezExperimenter

Generating code for LLM and VLM parallel runs (with or without structured JSON output)

Sdamirsa/GI-Copilot-v0.2

GI copilot is a multimodal RAG for answering to provide decision support to gastroenterologists.

Sdamirsa/LLM-VLM-in-Gastroenterology

Large Language Models (LLM), and Vision Language Models (VLM) performance in gastroenterology.

Sdamirsa/LLMs_in_GI

Papers, models, and datasets aiming at evaluation of Large Language Models (LLM) in gastroenterology

Sdamirsa/MedLSAM

MedLSAM: Localize and Segment Anything Model for 3D Medical Images

Sdamirsa/Merge-guideline-LLM-tool-Public

Merge-guideline-LLM-tool-Public

Sdamirsa/My_Full_stack_development_path

Sdamirsa/My_Saved_Codes

My saved codes

Sdamirsa/no-code-app-for-GPT-research

No code streamlit app for running GPT models on excel file (and ground truth). This could be used for (1) generating unstructured output, or (2) structured output (for extraction or generating variables).

Sdamirsa/PanCanAID-SAM-Home

PanCanAID SAM Home

Sdamirsa/PanCanAID-tools

The tools for managing PanCanAID project

Sdamirsa/PEPML

Sdamirsa/Sinopharm_C19_Response_HemodyalsisPatients

Sinopharm Vaccine Respone in Hemodyalisis Patients.

Sdamirsa/Stapp_in_colonochartreview_v1.0_stable

Stapp_in_colonochartreview_v1.0_stable

Sdamirsa/Tehran_COVID_Cohort_LLMExperiment

Sdamirsa/TiLense-4BlackBox-VLM

TiLense-4BlackBox-VLM is a tool designed to visualize and interpret the inner workings of vision-language models (VLMs). By generating heatmaps that highlight important image regions (tiles) based on model prediction behavior.

Sdamirsa/XNAT-OHIF_read-overlay-visualize-segmentation-file

This repo is a easy-to-use code which will read the segmentation file (.dcm) created by OHIF or XNAT, and overlay this with corresponding slice of corresponding series. Hope you can enjoy it.