AI model training based on MRI of 1,029 patients with brain metastases from a single institute and single MRI brand was conducted. A benchmark algorithm of 2D Mask R-CNN was used and resulted in an initial model (DeepMets®). Model generalization of DeepMets®was then carried out over a nationwide population-based dataset via deep active learning on 559 patients (randomized from 3,125) from National Health Insurance Administration (NHIA) medi-cloud, Taiwan. Iterative refining process using the ResNext50 U-Net architecture with attention mechanisms were undertaken and resulted in a newer version model (DeepMets®-Plus).
Taiwan Medical Imaging Co., Ltd. is dedicated to providing medical teams with validated AI medical imaging detection and predictive models for both clinics and wards. The company offers a complete usage system to assist physicians in disease detection, prognosis prediction, treatment tracking, and medical report generation more quickly and accurately. Additionally, Taiwan Medical Imaging provides hospitals with an open federated learning platform, allowing physicians, AI engineers, and data scientists from different institutions to collaboratively develop and validate cross-center AI models while safeguarding privacy and human rights, accelerating the adoption of AI-driven hospitals and precision healthcare worldwide.
Technology maturity：Mass production
Exhibiting purpose：Display of scientific results
Trading preferences：Technical license/cooperation、Negotiate by self
*Organization
*Name
*Phone
*Main Purpose
*Discuss Further
*Job Category
*Overall Rating
*Favorite Area
*Key Tech Focus
*Willing to Receive Updates?
Other Suggestions
Coming soon!