2020 № 4 Moscow experiment on computer vision in radiology: involvement and participation of radiologists
B a c k g r o u n d . In 2019, the Moscow Government decided to conduct a large-scale scientific research – an the Experiment on the use of innovative computer vision technologies for medical image analysis and subsequent applicability in the healthcare system of Moscow (www.mosmed.ai).
O b j e c t i v e – analyze engagement, attitudes and feedback from doctors-radiologists in frame of the Experiment.
M a t e r i a l s a n d m e t h o d s . The Experiment is a prospective research approved by the Independent Ethics Committee and registered with Clinicaltrails.gov (ID NCT04489992). Patients signed informed voluntary consent. On the date 01.10.2020, ten services are involved in the Experiment, they providing automated analysis of chest computed tomography and x-ray, mammography. The study includes quantitative indicators of the Experiment from 06/18/2020 to 10/01/2020. Methods of social survey, descriptive statistics, assessment of diagnostic accuracy metrics were used.
R e s u l t s a n d d i s c u s s i o n . During the first four months of the active phase of the Experiment, ten computer vision services were integrate into Unified Radiology Service of Moscow. More then 497 thousand studies have been successfully analyzed. Analyzes is carried out for 884 diagnostic devices in 293 medical organizations, 272 of them are actively involved. The involvement of medical organizations is 82%. The median time for automatic analysis of 1 study is 8 minutes. Overall, 63% of studies were analyzed in less than 15 minutes. At the beginning of the Experiment, 538 doctors had access to the system; in four months this number increased to 899. The involvement of doctors was 24%, which is slightly higher than the global indicators. According to the results of a sociological survey, the attitude to AI technologies of Moscow radiologists can be characterize as expectant, moderately optimistic. Radiologists
have determined that the results of computer vision services are fully consistent with the real situation in 64% of cases. In 36% cases some inconsistencies were recorded; of this number, significant discrepancies took place in 6%, insignificant – in 23%.
C o n c l u s i o n . Results of the Experiment’s first four months can be consider as successful. A high level of involvement of radiologists is define. Special measures will be implement to increase the involvement of radiologists, as well as a comprehensive comparative assessment of the work of services at the further stages of the Experiment.
2019 № 1 Benchmarking for assessing the quality of digitization of radiology departments: methodology development
Basing on the “Digital Imaging Adoption Model (DIAM)”, systematized international and own practical experience, there
was developed a tool for benchmarking the level of digitization of radiology departments. The tool includes: 1) a reference “Medviz” maturity model of medical imaging digitization; 2) a basic methodological document (unified five-level digitalization strategy); 3) a digitization degree assessment tool; 4) a structured set of applications. In process of benchmarking, radiology departments are classified in accordance with the author’s Medviz model. Following the established level, a special set of recommendations is formed as basis for further managerial decisions. The methodology was successfully tested by us in five medical centers of the Department of Healthcare of Moscow. The development is applicable for complex studies of IT support of medical centers.
2020 № 2 Telemedicine and COVID-19: quality of patient-initiated teleconsultations in case of acute respiratory disease
There is a global increase in demand for direct-to-patient telemedicine consultations due to COVID-19 pandemic. We made a quality assessment of the patient-initiated consultations in case of acute respiratory viral infection symptoms (COVID-19). There are 20 teleconsultations of 2 simulated patients in 10 the most popular telehealth services. An incomplete history of the disease was recorded in 50.0% of cases, incomplete allergy anamnesis – in 60.0%, and epidemiology anamnesis – in 35.0%. Information on chronic illnesses (critical for COVID-19 suspected situations) is fully collected only in 50.0% of cases. Due to defects in the history taking the target diagnostic concept was achieved in 30.0% of teleconsultations, target actions were recommended in only 35.0%. Telemedicine services did not provide continuity of medical care. In 60.0% of cases, medications were prescribed, including injectable antibacterial agents, which completely contradicts not only the legislation, but also the accepted international methods and practices of distance counseling. The quality of the direct-to-patient telemedicine services remains unsatisfactory. There is no effective quality control and quality assurance systems.
2020 № 1 The quality of primary direct-to-consumer telemedicine consultations (by results of testing telemedicine services)
The adoption of a number of special regulatory documents has significantly expanded the possibilities of using
telemedicine technologies in the Russian Federation. A significant number of services have appeared on the market, offering
direct-to-consumer and patient initiated telemedicine consultations. The explosive growth in the number of such services requires
a careful study of the telemedicine interaction quality. The research objective is to assess the quality of remote interaction of
participants in primary telemedicine consultations with simulated patients. For testing, we used descriptions of simulated patients
basing on real records of patients from the therapeutic department of the municipal clinical hospital. For objectification and
analysis of the telemedicine interaction process (collection of complaints, questioning, recommendations, etc.) a checklist has been compiled. Initially, the study included such services as “Yandex. Health”, “MMT (OnlineDoctor)”, “TelemedHelp”, “Doctor at work”,
“DoctorSmart”, “DocDoc”; then, for technical reasons, 2 services were excluded from the study. Two simulated patients were sent
to 4 services; as results, 8 primary teleconsultations were made. In all cases, incorrect, incomplete collection of anamnesis data
was recorded. Target diagnosis (in the form of diagnostic concept) was achieved in 25% of cases, and target prescriptions were
made in 50% of cases. A personal consultation and additional examinations were recommended in 75% of teleconsultations.
There was no continuity, drugs were somehow prescribed in 62.5% of cases. As results, an urgent development of methods of
internal and departmental quality control of medical care provided with the use of telemedicine technologies is required.
2019 № 2 Teleradiology in Russian Federation: state-of-art
2018 № 4 Expert Telemedicine Consultations in Radiology Service of Moscow
In 2017 Moscow Research and Practical Center of Medical Radiology conducted 2678 expert telemedicine consultations (i.e., TMC) following the requests from the radiology departments in Moscow municipal hospitals. A system-integrated base for distance consultations is called Unified Radiological Information Service (i.e., URIS). Expert TMCs were performed based on their modality, i.e. 52% (1386 cases) for MR imaging, 47% (1257 cases) for CT, 1% (35 cases) for X-ray imaging. The vast majority (i.e., 96.7% cases) of all applications were received from municipal outpatient clinics, providing healthcare assistance to the adult population. 3.2% of cases of expert TMC were conducted on an urgent basis. For the first time we defined the average demand for expert TMC results of radiology studies to be equal to 4,8 teleconsultations for 1,000 CTs, 17,42 – for 1 CT scanner, 8,9 – for 1,000 MRI, 26,98 – for 1 MRI scanner
2018 № 2 «Cross-reporting» as a way for sub-speciality teleradiology
There is an original telemedicinebased approach for a subspeciality reporting of a radiology examinations at primary level facilities. This approach allows to increase quality of «neuroradiology» reporting at children. The level of discrepancies decreased from 42,0% to 20,0%.