Inside the MATRIX — #1: How does our AI assisted cancer diagnosis and treatment system work? The National Health and Welfare Commission invited us to share details
https://medium.com/@matrixainetwork/inside-the-matrix-1-how-does-our-ai-assisted-cancer-diagnosis-and-treatment-system-work-19186363959This is the first of a brand new series to get a closer look at MATRIX. We think communication is important and we want to share all of the great things we’re achieving with our supporters.
Recently we were invited to The Pathology Quality Control Evaluation Center to visit the remote disease diagnosis and quality control platform, as well as the Center for Disease Diagnosis, to carry out technical exchanges and discuss cooperation.
The Pathology Quality Control Evaluation Center is a digital remote diagnosis and quality control platform for digital pathology established by the National Health and Welfare Commission of the People’s Government of the People’s Republic of China. The Center integrates various resources and the use of modern network technologies and digital slice image processing systems to achieve remote diagnosis goals.
The Center informed us that the current medical and disease research is developing rapidly, and that although pathologists are experienced, it is difficult to diagnose the diseases encountered in a short period of time. It is hoped that the introduction of new technologies will provide more space for quick and accurate pathological diagnoses.
Clinical diagnoses are made according to identifying common signs and patterns based on past information and images based on accumulated experience. AI can copy the whole process and learn quickly and more efficiently, whilst being able to recall the information more readily when it’s needed.
Through deep learning technology, AI can continue to learn X-rays, medical books, essays, electronic medical records, and things like the formation of tumor tissue and normal histopathology; anything which is already available in huge databases. When a digital pathological section needs to be diagnosed, the most similar sections are quickly retrieved from the database by image recognition technology and diagnosed to help the doctor formulate a treatment plan.
AI-assisted disease diagnosis and treatment is an inevitable future trend of medical development. It is the perfect combination of pathological image recognition technology and machine learning methods, and the deep learning facilities provided by the our bayesian Proof of Work algorithm is a perfect training ground. Not only will it help with Pathologists’ day-to-day work which requires high degrees of concentration and an extremely wide knowledge base, it also facilitates a greatly increased accuracy and efficiency of pathological diagnosis.
The diagnosis and treatment solution we are currently building is one to aid in the diagnosis of thyroid cancer and liver cancer. We are working with Beijing Cancer Hospital, the 302 Hospital and a number of other first-line hospitals in the China to realize this goal — our solution will be in place in each of these hospitals when it’s ready.
In the next phase, we will continue to strengthen communication with The Center, work to improve the AI technology used in digital pathological diagnosis with the hospitals, optimize the cooperation details and promote the common development of both parties.
