Jooyoung Lee, Woo Sang Cho, Byeong Soo Kim, Dan Yoon, Jung Kim, Ji Hyun Song, Sun Young Yang, Seon Hee Lim, Goh Eun Chung, Ji Min Choi, Yoo Min Han, Hyoun-Joong Kong, Jung Chan Lee, Sungwan Kim, Jung Ho Bae. Gut and Liver 2024;18:857-66. https://doi.org/10.5009/gnl240068

Background: Identifying patients unlikely to achieve meaningful improvement following total knee arthroplasty (TKA) supports more effective shared decision-making (SDM). This study aimed to develop and validate machine learning (ML) models that preoperatively predict insufficient clinical improvement one year after TKA using Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) subscales and total scores, and to assess the important predictive variables. Methods: A retrospective analysis was performed on consecutive primary TKA patients from 2004 to 2022 at a single tertiary hospital was conducted. Insufficient clinical improvement was defined as not achieving the minimal clinically important difference (MCID) for each WOMAC subscale and total. Candidate preoperative variables included demographics, comorbidities, knee range of motion, radiologic variables, and WOMAC scores. A variety of ML models were evaluated using performance metrics for calibration and discrimination, as well as decision curve analysis and Shapley additive explanations. Results: Among the 3,810 TKAs included, the ExtraTrees model performed best for WOMAC pain, stiffness, function, and total scores, achieving AUCs of 0.92, 0.90, 0.87, and 0.89; recall rates of 0.79, 0.86, 0.70, and 0.83; and Brier scores of 0.09, 0.10, 0.11, and 0.06, respectively, along with demonstrating good calibration curves and net clinical benefit. Shapley additive explanations identified better preoperative WOMAC scores, osteoporosis, diabetes mellitus, older age, malignancy, and coronary artery disease as important predictors of insufficient clinical improvement. Conclusions: The ML models demonstrated good performance in preoperatively predicting insufficient clinical improvement at 1 year after TKA based on WOMAC. These models have the potential to enhance SDM and perioperative patient management by preoperatively identifying approximately 70% to over 80% of patients likely to experience insufficient clinical improvement, with a specificity of about 80%, and by providing explanations regarding associated factors.

Robotic surgery is preferred over open or laparoscopic surgeries due to its intuitiveness and convenience. However, prolonged use of surgical robots can cause neck pain and joint fatigue in wrist and fingers. Also, input systems are bulky and difficult to maintain. To resolve these issues, we propose a novel input module based on real-time 3D hand tracking driven by RGB images and MediaPipe framework to control surgical robots such as patient side manipulator (PSM) and endoscopic camera manipulator (ECM) of da Vinci research kit. In this paper, we explore the mathematical basis of the proposed 3D hand tracking module and provide a proof-of-concept through user experience (UX) studies conducted in a virtual environment. End-to-end latencies for controlling PSM and ECM were 170 ± 10 ms and 270 ± 10 ms, respectively. Of fifteen novice participants recruited for the UX study, thirteen managed to reach a qualifiable level of proficiency after 50 min of practice and fatigue of hand and wrist were imperceivable. Therefore, we concluded that we have successfully developed a robust 3D hand tracking module for surgical robot control and in the future, it would hopefully reduce hardware cost and volume as well as resolve ergonomic problems. Furthermore, RGB image driven 3D hand tracking module developed in our study can be widely applicable to diverse fields such as extended reality (XR) development and remote robot control. In addition, we provide a new standard for evaluating novel input modalities of XR environments from a UX perspective.

Sensors in the medical field to detect specific tissues, such as radars, must provide accurate signals from frequency generators. In order to supply an accurate frequency signal, the oscillator must have a low phase noise. Therefore, the resonator used in the oscillator must provide a high Q_L. Therefore, in this paper, we have proposed a low-phase-noise X-band oscillator that used a resonator with a high value of Q_L as a sensor for tissue-locating applications. The resonator had a split-ring structure and consisted of an open-loaded, T-type stub with a high-Q_L; such high-Q_L levels were enabled by controlling the length of the open-circuit in the T-type stub. This led to the generation of only low-phase noise in the proposed oscillator. Experimental results showed that, at an operating frequency of 10.08 GHz, the output power was 18.66 dBm, the second harmonic suppression was −34.40 dBc, and the phase noise was −138.13 dBc/Hz at an offset of 100 kHz. This proposed oscillator can be used as a sensor to detect the location of tissues during laparoscopic surgery.