Jae‐Seung Jeong

Hyungjun Park, Chang-Yun Woo, Seungjo Lim et al.

Objective To develop an LLM based realtime compound diagnostic medical AI interface and performed a clinical trial comparing this interface and physicians for common internal medicine cases based on the United States Medical License Exam (USMLE) Step 2 Clinical Skill (CS) style exams. Methods A nonrandomized clinical trial was conducted on August 20, 2024. We recruited one general physician, two internal medicine residents (2nd and 3rd year), and five simulated patients. The clinical vignettes were adapted from the USMLE Step 2 CS style exams. We developed 10 representative internal medicine cases based on actual patients and included information available on initial diagnostic evaluation. Primary outcome was the accuracy of the first differential diagnosis. Repeatability was evaluated based on the proportion of agreement. Results The accuracy of the physicians' first differential diagnosis ranged from 50% to 70%, whereas the realtime compound diagnostic medical AI interface achieved an accuracy of 80%. The proportion of agreement for the first differential diagnosis was 0.7. The accuracy of the first and second differential diagnoses ranged from 70% to 90% for physicians, whereas the AI interface achieved an accuracy rate of 100%. The average time for the AI interface (557 sec) was 44.6% shorter than that of the physicians (1006 sec). The AI interface ($0.08) also reduced costs by 98.1% compared to the physicians' average ($4.2). Patient satisfaction scores ranged from 4.2 to 4.3 for care by physicians and were 3.9 for the AI interface Conclusion An LLM based realtime compound diagnostic medical AI interface demonstrated diagnostic accuracy and patient satisfaction comparable to those of a physician, while requiring less time and lower costs. These findings suggest that AI interfaces may have the potential to assist primary care consultations for common internal medicine cases.

Jaehyun Kim, Jaeseung Jeong

Despite advances in hierarchical reinforcement learning, its applications to path planning in autonomous driving on highways are challenging. One reason is that conventional hierarchical reinforcement learning approaches are not amenable to autonomous driving due to its riskiness: the agent must move avoiding multiple obstacles such as other agents that are highly unpredictable, thus safe regions are small, scattered, and changeable over time. To overcome this challenge, we propose a spatially hierarchical reinforcement learning method for state space and policy space. The high-level policy selects not only behavioral sub-policy but also regions to pay mind to in state space and for outline in policy space. Subsequently, the low-level policy elaborates the short-term goal position of the agent within the outline of the region selected by the high-level command. The network structure and optimization suggested in our method are as concise as those of single-level methods. Experiments on the environment with various shapes of roads showed that our method finds the nearly optimal policies from early episodes, outperforming a baseline hierarchical reinforcement learning method, especially in narrow and complex roads. The resulting trajectories on the roads were similar to those of human strategies on the behavioral planning level.

Seong-il Im, Hyejin Lee, Jaesang Lee et al.

Interest in Restricted Boltzmann Machine (RBM) is growing as a generative stochastic artificial neural network to implement a novel energy-efficient machine-learning (ML) technique. For a hardware implementation of the RBM, an essential building block is a reliable stochastic binary neuron device that generates random spikes following the Boltzmann distribution. Here, we propose a highly-scalable stochastic neuron device based on Ovonic Threshold Switch (OTS) which utilizes the random emission and capture process of traps as the source of stochasticity. The switching probability is well described by the Boltzmann distribution, which can be controlled by operating parameters. As a candidate for a true random number generator (TRNG), it passes 15 among the 16 tests of the National Institute of Standards and Technology (NIST) Statistical Test Suite (Special Publication 800-22). In addition, the recognition task of handwritten digits (MNIST) is demonstrated using a simulated RBM network consisting of the proposed device with a maximum recognition accuracy of 86.07 %. Furthermore, reconstruction of images is successfully demonstrated using images contaminated with noises, resulting in images with the noise removed. These results show the promising properties of OTS-based stochastic neuron devices for applications in RBM systems.