Disease monitoring
In the field of disease surveillance, doctors can use a predictive model based on machine learning or cognitive systems to estimate the risk of chronic diseases based on the patient's characteristics, without sticking to an established care plan or allowing patients to repeat hospital admissions. Early intervention can significantly reduce patient costs. Montefiore Health System deploys a data analysis platform based on the Intel® Xeon® processor, which analyzes large amounts of raw data in real time to help clinicians determine the best treatment for patients. Planning. At the same time, the normative model can also be used to identify the risk of respiratory failure in patients so that medical personnel can take early warning measures to intervene in time to save lives and conserve resources.
Clinical environment
Machine learning based models can also be used in the clinical environment. Common predictive models include the use of electronic medical record data to assess the risk of infection in hospitals, the use of models to predict the probability of patients entering the accident room, etc. Intel and Sharp Healthcare jointly develop The rapid-response team model can predict which patients need the intervention of the rapid response team based on the data in the electronic medical records. At the same time, the hospital can quickly find the corresponding emergency personnel and equipment, thereby shortening the response time. In the use of historical data on In the experiments conducted by the model, the accuracy of estimating the patient's need for rapid reaction group intervention is about 80%.
Imaging Analysis
The use of deep learning to analyze medical images is also one of the important applications of artificial intelligence technology in the medical field. In this regard, Intel has collaborated with industry partners to analyze medical images using in-depth learning technology for tumor detection. In cooperation with GE Healthcare, GE Healthcare uses the Intel® Xeon® Scalable Platform to reduce the total cost of ownership of imaging equipment by 25%. By working with GE Healthcare's imaging solutions, the Intel Xeon Extensible Platform can help radiologists improve their reading efficiency. The first image display time was reduced to less than 2 seconds, and all study load time was reduced to less than 8 seconds.
Virtual service
The fourth use case of artificial intelligence is a virtual service represented by telemedicine. The application of telemedicine provides a richer solution for businesses and consumers. InTouch Health, a hospitalized medical robot, is one of the representatives of novel solutions. The resulting video dataset can be used to develop artificial intelligence solutions to further improve clinical diagnosis. For example, in the case of remote stroke disease diagnosis, the model based on deep learning can identify early stroke characteristics of patients, and then increase the diagnostic accuracy rate. Significantly shorten the time of diagnosis and treatment.
Virtual Reality
The fifth use case of artificial intelligence is to create next-generation virtual reality assistants. In the future, artificial intelligence can respond to participants' interactions in virtual reality sessions. Patients can interact with virtual environments and observe possible changes in their condition. In terms of surgical training, artificial intelligence can be used to analyze images to identify top surgeons' best practices. These methods can be fed back into simulations and can be improved over time.
In summary, digital transformation has brought new opportunities in the medical and health field. In the process of transformation, medical organizations should use data as a core capability to enhance business processes and patient experience. With the further improvement of computational analysis capabilities, artificial intelligence is in the medical and health field. Field application scenarios will be more abundant.