Working with Huawei, Shanghai Ruijin Hospital open source RuiPath Path Model – Passionategeekz

Passionategeekz On June 30, Shanghai Ruijin Hospital joined hands with Huawei to open the source of the core visual basic model in the RuiPath pathological model, which covers seven major high-incidence cancers, including lung cancer and colorectal cancer.Covering 90% of new cancer cases in China each year。

According to reports, the model was published on February 18thbased on Huawei DCS AI solution, this open source content includes core visual architecture and multiple cancer test data sets, marking an important breakthrough in my country’s digital pathological AI field. As the first open source pathology model for medical institutions in Shanghai, RuiPath adopts cutting-edge deep learning technology to achieve high-precision pathological image analysis.

This model can help doctors improve diagnosis efficiency. Ruijin Hospital pathologist Jason Qian said: “This model covers 90% of the cancer-occurring population in China every year; in terms of depth, the sub-specialist knowledge question-and-answer level has reached the level of expert knowledge. RuiPath’s answers in common question tests compiled by pathologistsAccuracy rate is as high as 90% or moreand is at the leading level at home and abroad in the graphic and text Q&A tasks in medical examination scenarios. ”

After checking the official website of Huawei’s enterprise business, Passionategeekz learned that DCS AI solutions can make data engineering tools in pathological scenarios in the medical industry.Shorten the medical training data preparation cycle 80%;Its ModelEngine has built-in 40+ special data processing operators, which can achieve rapid generation of medical knowledge.

At the same time, the solution also has a unique CSP data preprocessing patch-free slicing operator, which allows preprocessing of millions of slices to be performed on a preprocessing time of millions of slicesFrom Month to Heavenly Levelthe overall system-level model training and inference acceleration capability of this solution, achieving a 30% reduction in model training cycle and double inference concurrency.

Advertising statement: The external redirect links (including, not limited to, hyperlinks, QR codes, passwords, etc.) contained in the article are used to convey more information and save selection time. The results are for reference only. All articles from Passionategeekz include this statement.