Shin TY, et al. Investig Clin Urol. 2020 Nov;61(6):555-564. https://doi.org/10.4111/icu.20200086

Park H, et al. J Korean Med Sci. 2018 Jan;33(11):e94. https://doi.org/10.3346/jkms.2018.33.e94

This study aims to compare the effectiveness of using discrete heartbeats versus an entire 12-lead electrocardiogram (ECG) as the input for predicting future occurrences of arrhythmia and atrial fibrillation using deep learning models. Experiments were conducted using two types of inputs: a combination of discrete heartbeats extracted from 12-lead ECG and an entire 12-lead ECG signal of 10 s. This study utilized 326,904 ECG signals from 134,447 patients and categorized them into three groups: true–normal sinus rhythm (T-NSR), atrial fibrillation–normal sinus rhythm (AF-NSR), and clinically important arrhythmia–normal sinus rhythm (CIA-NSR). The T-NSR group comprised patients with at least three normal rhythms in a year and no atrial fibrillation or arrhythmias history. Clinically important arrhythmia included atrial fibrillation, atrial flutter, atrial premature contraction, atrial tachycardia, ventricular premature contraction, ventricular tachycardia, right and left bundle branch block, and atrioventricular block over the second degree. The AF-NSR group included normal sinus rhythm paired with atrial fibrillation or atrial flutter within 14 days, and the CIA-NSR group comprised normal sinus rhythm paired with CIA occurring within 14 days. Three deep learning models, ResNet-18, LSTM, and Transformer-based models, were utilized to distinguish T-NSR from AF-NSR and T-NSR from CIA-NSR. The experiments demonstrated the potential of using discrete heartbeats in predicting future arrhythmia and atrial fibrillation incidences extracted from 12-lead electrocardiogram (ECG) signals alone, without any additional patient information. The analysis reveals that these discrete heartbeats contain subtle patterns that deep learning models can identify. Focusing on discrete heartbeats may lead to more timely and accurate diagnoses of these conditions, improving patient outcomes and enabling automated diagnosis using ECG signals as a biomarker.

Park WH, et al. J Korean Med Sci. 2014 Jul;29(7):1018-1020. https://doi.org/10.3346/jkms.2014.29.7.1018

Shin JH, et al. Investig Clin Urol. 2023 May;64(3):255-264. https://doi.org/10.4111/icu.20220411

To evaluate the clinical utility of percentage of serum prostate-specific antigen (proPSA) to free PSA (%p2PSA) and the prostate health index (PHI) for predicting aggressive pathological outcomes of radical prostatectomy (RP) in Korean males. This prospective observational multicenter study included 160 Korean males who consecutively underwent RP. The predictive utility of preoperative %p2PSA and PHI for predicting the following pathological outcomes of RP including pT3 disease, pathologic Gleason sum ≥7, and Gleason sum upgrading was investigated using multivariate and decision-curve analyses. The PHI and %p2PSA levels were significantly higher in patients with pT3 disease, pathologic Gleason sum ≥7, and Gleason sum upgrading. On univariate analysis, PHI was an accurate predictor of pT3 disease, pathologic Gleason sum ≥7, and Gleason sum upgrading. Multivariate and decision curve analyses revealed that inclusion of PHI to a base multivariate model including total PSA, percentage free PSA, PSA density, percentage of positive biopsy core, biopsy Gleason sum, and clinical stage factors significantly increased its predictive accuracy; %p2PSA showed a similar result. However, PHI was a more valuable predictor of pathological outcomes of RP. This study revealed PHI and %p2PSA as preoperative biomarkers of pathological outcomes in Korean males who underwent RP for prostate cancer.