The DeepCT is the world’s leading AI-driven system developed by Deep01 to identify abnormal CT scan images. Recently, Kaohsiung Veterans General Hospital (KVGH), the largest hospital in southern Taiwan, as the third-party medical center released clinical trial results of using the DeepCT system in its emergency department. The trial including a total of 2,999 patients shows that the use of DeepCT shortened the intracranial hemorrhage (ICH) patients’ length of stay (LOS) at the emergency department by 3 hours and 40 minutes on average. This means the DeepCT system will greatly reduce the ICH patients’ waiting time in the emergency department and accelerate the follow-up treatment procedure, which is immensely beneficial to both of the patients and the hospital management. The clinical trial results were accepted and published in June this year by the high-impact-factor Journal of Medical System in the US.
ICH is a common clinical condition in the emergency department. It’s usually caused by traumatic brain injuries or strokes. For ICH patients, time is crucial, and the medical staff have to race against time to make accurate decisions based on diagnosis within a very limited amount of time. The DeepCT AI system has an accuracy rate of up to 99.16% covering in the clinical trial 238 patients with 96.43% sensitivity and 99.52% specificity. It’s worth noting that in the same study, 2999 patients were counted, of whom the ICH patients ended up having LOS reduced, upon the implementation of the DeepCT system, from 781 minutes to 561 minutes, a decrease of 3 hours and 40 minutes, namely, by 28%. This will accelerate ICH patients’ follow-up treatments and help with their health recovery later on.
Yang Tsung-Lung, Director of KVGH’s Quality Management Center holds the view that KVGH is proposing a “Three Appropriates” solution to respond to the time factor in severe acute medical treatment, namely, offering the appropriate treatment to the appropriate patients at the appropriate time. The AI-assisted ICH imaging interpretation by the DeepCT is exactly what is needed in a time-critical medical emergency. KVGH’s quality management center section chief and the research project’s principal investigator, Juang Wang-Chuan, stated that regarding the DeepCT ICH detection system, the initial focus was largely on the accuracy of AI-assisted ICH interpretation, However, during the execution of the research project, it was noted that the ICH patients’ LOS was reduced. This means the ICH patients will be fast forwarded to the next stage of the treatment, be it hospitalization, ICU or the surgical room. This is absolutely great news to the patients and beneficial to the hospital’s management quality improvement, and as a result, implementing the concept of “Lean Management” and reducing time wasted on waiting for the imaging interpretation.
The fact that these results and data were published in a high-impact-factor journal shows that AI is now being acknowledged for its effectiveness in medical treatment, according to Dr. Chen Chih-Yu of the Ministry of Health and Welfare Shuangho Hospital. Arthur Chen, who took part in this research project also stated that Deep01‘s AI-assisted system will reduce the patient’s waiting time and the time for doctors’ clinical decision-making; more importantly, it further verifies that AI-driven products will bring to the hospital concrete and quantified economic value and will serve as reference for insurance payouts. Taiwan’s AI-supported health care is up there with the world’s best, and Deep01 is one of the contributing factors.