Fujitsu AI Tool Enhances Genomic Medicine, Cancer Treatment Planning

Fujitsu has developed explainable AI technologies for use in genomic medicine and cancer treatment planning.

The Japanese technology firm unveiled an AI-powered tool that creates knowledge graphs from large-scale data sets.

By bringing together various data sources, including text, images and numerical data, Fujtisu’s new AI system helps scientists developing therapeutics to interpret vast troves of data, by highlighting key insights.

The AI is designed to support human researchers by highlighting visual cues and patterns, making disease classification easier and expediting treatment planning.

Fujitsu also unveiled a technology capable of training algorithms on features in medical images to help with diagnoses.

Credit: Fujitsu

Fujitsu tested the tools on medical benchmarks including lung cancer type classification and breast cancer patient survival prediction.

The AI systems demonstrated the capability to analyze genomic and pathological data to classify different types of lung cancer, with 92.1% accuracy and outperformed previous analysis tools.

The AI could also accurately predict a patient’s survival predictions by analyzing image data and ribonucleic Acid (RNA) information, achieving a score of 71.8%, higher than the previous best accuracy of 66.8%.

“These tests confirmed that Fujitsu’s technology can accurately support the identification of two main types of lung cancer, for example, by illuminating the factors behind the pathological classification based on key visual cues,” according to a company announcement.

Related:FDA Clears AI Tool for Detecting Cancer Signs in Bone Marrow

Fujitsu plans to expand development of its new multimodal analysis tools for a wider range of medical disciplines. By the end of 2024, the company plans to make them available through its Fujitsu Research Portal, where researchers can test its AI solutions.