Drowning diagnosis is a complicated process in the autopsy, even with the assistance of autopsy imaging and the on-site information from where the body was found. Previous studies have developed well-performed deep learning (DL) models for drowning diagnosis. However, the validity of the DL models was not assessed, raising doubts about whether the learned features accurately represented the medical findings observed by human experts. In this paper, we assessed the medical validity of DL models that had achieved high classification performance for drowning diagnosis. This retrospective study included autopsy cases aged 8-91 years who underwent postmortem computed tomography between 2012 and 2021 (153 drowning and 160 non-drowning cases). We first trained three deep learning models from a previous work and generated saliency maps that highlight important features in the input. To assess the validity of models, pixel-level annotations were created by four radiological technologists and further quantitatively compared with the saliency maps. All the three models demonstrated high classification performance with areas under the receiver operating characteristic curves of 0.94, 0.97, and 0.98, respectively. On the other hand, the assessment results revealed unexpected inconsistency between annotations and models' saliency maps. In fact, each model had, respectively, around 30%, 40%, and 80% of irrelevant areas in the saliency maps, suggesting the predictions of the DL models might be unreliable. The result alerts us in the careful assessment of DL tools, even those with high classification performance.

• MeSH
• Deep Learning * MeSH
• Child MeSH
• Adult MeSH
• Middle Aged MeSH
• Humans MeSH
• Adolescent MeSH
• Young Adult MeSH
• Autopsy * methods   MeSH
• Tomography, X-Ray Computed * methods   MeSH
• Postmortem Imaging MeSH
• Reproducibility of Results MeSH
• Retrospective Studies MeSH
• ROC Curve MeSH
• Aged, 80 and over MeSH
• Aged MeSH
• Drowning * diagnosis   MeSH
• Check Tag
• Child MeSH
• Adult MeSH
• Middle Aged MeSH
• Humans MeSH
• Adolescent MeSH
• Young Adult MeSH
• Male MeSH
• Aged, 80 and over MeSH
• Aged MeSH
• Female MeSH
• Publication type
• Journal Article MeSH

INTRODUCTION: We aim to compare the acute and long-term success of defibrillation between non-apical and apical ICD lead position. METHODS AND RESULTS: The position of the ventricular lead was recorded by the implanting physician for 2,475 of 2,500 subjects in the Shockless IMPLant Evaluation (SIMPLE) trial, and subjects were grouped accordingly as non-apical or apical. The success of intra-operative defibrillation testing and of subsequent clinical shocks were compared. Propensity scoring was used to adjust for the impact of differences in baseline variables between these groups. There were 541 leads that were implanted at a non-apical position (21.9%). Patients implanted with a non-apical lead had a higher rate of secondary prevention indication. Non-apical location resulted in a lower mean R-wave amplitude (14.0 vs. 15.2, P < 0.001), lower mean pacing impedance (662 ohm vs. 728 ohm, P < 0.001), and higher mean pacing threshold (0.70 V vs. 0.66 V, P = 0.01). Single-coil leads and cardiac resynchronization devices were used more often in non-apical implants. The success of intra-operative defibrillation was similar between propensity score matched groups (89%). Over a mean follow-up of 3 years, there were no significant differences in the yearly rates of appropriate shock (5.5% vs. 5.4%, P = 0.98), failed appropriate first shock (0.9% vs. 1.0%, P = 0.66), or the composite of failed shock or arrhythmic death (2.8% vs. 2.3% P = 0.35) according to lead location. CONCLUSION: We did not detect any reduction in the ICD efficacy at the time of implant or during follow-up in patients receiving a non-apical RV lead.

• MeSH
• Time Factors MeSH
• Defibrillators, Implantable * MeSH
• Electric Countershock adverse effects   instrumentation   methods   mortality   MeSH
• Electrophysiologic Techniques, Cardiac MeSH
• Kaplan-Meier Estimate MeSH
• Cardiac Pacing, Artificial MeSH
• Middle Aged MeSH
• Humans MeSH
• Logistic Models MeSH
• Death, Sudden, Cardiac etiology   MeSH
• Proportional Hazards Models MeSH
• Prospective Studies MeSH
• Prosthesis Design MeSH
• Risk Factors MeSH
• Prosthesis Failure MeSH
• Aged MeSH
• Arrhythmias, Cardiac complications   diagnosis   mortality   therapy   MeSH
• Propensity Score MeSH
• Treatment Outcome MeSH
• Check Tag
• Middle Aged MeSH
• Humans MeSH
• Male MeSH
• Aged MeSH
• Female MeSH
• Publication type
• Journal Article MeSH
• Multicenter Study MeSH
• Research Support, Non-U.S. Gov't MeSH
• Randomized Controlled Trial MeSH
• Comparative Study MeSH