MIPT: MeMo: Meeting Moments AI Assistant

The main articles are:

• Speech Recognition (Technology, Market)
• Speech technology: On the path from recognition to understanding
• Artificial Intelligence in Medicine
• Virtual Assistants

2024: MeMo AI Assistant Announcement: Meeting Moments

MIPT taught artificial intelligence In to discharge sick leave. Recognising the doctor's voice in real time, AI the assistant to MeMo: Meeting Moments not only keeps an appointment record, but also fills out medical documentation: a patient's personal card, referrals to narrow specialists or a prescription. The solution proposed by students of the Landau School of Physics and Research, as well as the Department of BFC MIPT, will reduce the time that the doctor spends on non-core tasks by more than a third. Also, the use of automation will increase by 20% the number of patients that the doctor takes per working day. This was announced on August 13, 2024 by representatives of the Moscow Institute of Physics and Technology.

source = MIPT

According to statistics, the doctor spends up to 40% of the patient's admission time on reporting: filling out documentation, writing out referrals, etc., which negatively affects the quality of the medical services provided. The doctor's attention is not focused on the patient, appointments are delayed, queues are formed. As a result, the clinic (medical center) suffers economic losses. We want to change the rules of the game, according to which doctors are forced to conduct medical examinations and consultations, allowing specialists not to be distracted from patients. The medical record will be kept by algorithms. told Maria Protopopova

According to the company, one of the priority areas of the National Strategy for the Development of Artificial Intelligence (AI) in the Russian Federation is the development of a market for AI-based software products for healthcare. Moreover, as of August 2024, the state is interested not only in the development of algorithms for the processing of medical statistics, computer modeling of medicines, but also in the development of a patient-oriented culture. The latter, according to the author of the MeMo: Meeting Moments project, Maria Protopopova, is impossible without the introduction of automation into the medical admission procedure. This is the only way to free up the doctor's intellectual resource from solving routine tasks.

The MeMo: Meeting Moments development team has already completed major software options. The AI assistant can convert speech into text in real time and make summaries of information - fill out preloaded templates of reception protocols. Also, the startup team, in addition to developing technology, conducted castdevs and received support from the Moscow Center for Innovative Technologies in Healthcare (Medtech.Moscow). The authors of the project clarify that automation of admission will not violate the norms of medical secrecy.

source = MIPT

The doctor's dialogue with the patient and the contents of the medical record are confidential information, which, according to 152 Article FZ, is a "medical secret." This is one of the biggest challenges we have faced in working on a project. After analyzing the market and cases of similar solutions both in our country and abroad, we chose a business model in which the solution is fully installed on the servers in the data center (Data Center), on which the medical institution already operates. Our solution is software (application) installed on a computer in the doctor's office. And the processing of all data is just happening on the servers in the data center. Thus, all information remains in the contour of each clinic. The solution is fully local algorithms, without the use of third-party APIs (connections or servers). clarified Maria Protopopova

The MeMo: Meeting Moments project became the winner of the FiztekhIdea MIPT accelerator, and also received grant support from the Innovation Promotion Fund. The total amount of state investment in the startup of MIPT students amounted to 1.2 million rubles. By the end of August 2024, the project team plans to complete work on the local version of the application and will conduct pilot testing in one of the Moscow private clinics.