BM Smear AI 1.0
BM Smear AI 1.0 is an application within the CarbConnect platform. BM Smear AI 1.0 is an AI-powered research tool designed to automate the classification of bone marrow smear samples.
Trained on a large dataset of images captured using the CELLAVISION DC-1, this tool utilizes deep learning algorithms to efficiently detect and classify various bone marrow cell types.
Intended Uses and Limitations
This tool is strictly intended for Research Use Only (RUO).
It is not intended for:
- clinical diagnostics
- medical procedures
- patient management
- therapeutic selection
- any regulated medical use
For further details, please refer to the BM Smear Learn More page on CarbConnect.
How to Use
A video tutorial demonstrating how to use the app is available on YouTube.
Training Data
A training dataset was created using bone marrow smear specimens from 188 patients.
The dataset includes:
- 12,916 images
- 15,802 cells
- images captured using a CellaVision DC-1
- annotations for cell locations
- classification into 13 cell categories
Task
BM Smear AI 1.0 performs image-based classification of bone marrow smear cells for research workflows.
Notes
This repository documents the application and its intended research scope. It does not constitute clinical validation or regulatory clearance.