Submission for Google Gen AI Hackathon

Breast cancer histopathology image segmentation using U-Net. This repository implements U-Net for accurate segmentation of cancerous regions. It includes data augmentation, mixed precision training, checkpointing, and evaluation metrics like Dice score to improve model performance.

[Advanced] - [Machine Learning] - Focus on optimizing, make arguments about CNNs architecture and training-process of CNNs

Laboratory Catalog and Archive Service (LabCAS) Backend

These projects are a collection of the research I've done over the summer on utilizing a (somewhat) new math methodology know as the "randomized SVD" to see how it could be applied to issues such as cancer detection, facial, recognition, music recommendation, etc!

Pre-Processing Segmentation Datasets for the Hydra Framework. Contains preprocessing scripts and a module capable of creating segmentation masks

Convolutional neural network capable of identifying skin lesions (based on the skin lesion image data set HAM10000).

Machine Learning module for cloud-based analyses developed as part of the NIGMS Sandbox Project

A technological solution is sought that enables individuals to assess their own skin lesions through automated image analysis, thereby facilitating early detection and improving patient outcomes.

A CNN model to distinguish whether a skin lesion is malignant or benign: cancerous or not. Trained on 1500 images for both benign and malignant classes attained from the ISIC skin condition dataset with over 84% accuracy and 93 F1 score.

Detecting various characteristics of glioblastoma using Deep Learning

Exploring the revolutionary impact of Convolutional Neural Networks (CNNs) in detecting critical diseases such as lung, breast, and skin cancer, pneumonia, and COVID-19 through a systematic review of deep learning algorithms in medical imaging.

Official repository of "Towards Learning Contrast Kinetics with Multi-Condition Latent Diffusion Models"

This project leverages a refined U2Net for medical image segmentation, focusing on cancer cell analysis and proportion calculation to aid in diagnosis and treatment.

scMalignantFinder is a Python package specially designed for analyzing cancer single-cell RNA-seq datasets to distinguish malignant cells from their normal counterparts.

This repository contains various bioinformatics pipelines that explore different cancer types.

DeepHealth Annotate is a web-based tool for viewing and annotating DICOM images. Annotation metadata can be exported in JSON format to be used for a variety of purposes, such as creating training input for deep learning models that use bounding box algorithms.

🩺 Herramienta de análisis de imágenes médicas basada en 🤖 inteligencia artificial para la 🕵️‍♂️ detección temprana de cáncer.

A comprehensive classification tool based on pure transcriptomics for precision medicine

Cancer Image Detection With PyTorch