26 April 2024 in Biomedical Journal
Diagnostic imaging is essential in modern trauma care for initial evaluation and identifying injuries requiring intervention. Deep learning (DL) has become mainstream in medical image analysis and has shown promising efficacy for classification, segmentation, and lesion detection. This narrative review provides the fundamental concepts for developing DL algorithms in trauma imaging and presents an overview of current progress in each modality. DL has been applied to detect free fluid on Focused Assessment with Sonography for Trauma (FAST), traumatic findings on chest and pelvic X-rays, and computed tomography (CT) scans, identify intracranial hemorrhage on head CT, detect vertebral fractures, and identify injuries to organs like the spleen, liver, and lungs on abdominal and chest CT. Future directions involve expanding dataset size and diversity through federated learning, enhancing model explainability and transparency to build clinician trust, and integrating multimodal data to provide more meaningful insights into traumatic injuries. Though some commercial artificial intelligence products are Food and Drug Administration-approved for clinical use in the trauma field, adoption remains limited, highlighting the need for multi-disciplinary teams to engineer practical, real-world solutions. Overall, DL shows immense potential to improve the efficiency and accuracy of trauma imaging, but thoughtful development and validation are critical to ensure these technologies positively impact patient care.
1 April 2019 in The American Journal of Emergency Medicine
2 May 2017 in World Journal of Emergency Surgery
Emergent pancreaticoduodenectomy is a life-saving procedure in certain clinical scenarios when all the conservative treatment fails. The indications can be limited into perforation and bleeding. To clarify the impact of etiology on surgical outcomes of emergent pancreaticoduodenectomy for non-trauma, we analyzed our patients and performed a literature review. We reviewed 931 consecutive pancreaticoduodenectomies performed at our institute between January 2001 and July 2015. Patients with emergent pancreaticoduodenectomy for non-trauma etiologies were enrolled, whereas those who suffered from caustic injuries were excluded. The keywords “emergent/emergency” and “pancreaticoduodenectomy/pancreatoduodenectomy” were applied in a literature search. The universally available data for all the enrolled patients including etiology, surgical complications, outcomes, and hospital stays were analyzed. Univariate and multivariate logistic analysis for the contributing factors to surgical mortality were performed. Six out of 931 (0.6%) registered pancreaticoduodenectomies matched our criteria of inclusion. The literature review obtained 4 series and 7 case reports, which when combined with our patients yielded a cohort of 31 emergent pancreaticoduodenectomies with 13 cases of perforation and 18 of bleeding. The rate of emergent pancreaticoduodenectomy for non-traumatic etiologies is similar between the present study and the other 3 series, ranging from 0.3 to 3%. The overall surgical complication rate was 83.9%. The rate of surgical mortality is significantly higher than in elective pancreaticoduodenectomy by propensity score matching with age and gender (19.4 versus 3.2%, P = 0.015). Univariate and multivariate logistic regression disclosed that etiology is the only preoperative risk factor for surgical mortality (perforation versus bleeding; odds ratio = 39.494, P = 0.031). Emergent pancreaticoduodenectomy remains a rare operation. Surgical morbidity and mortality are higher than with elective pancreaticoduodenectomy among different reported series. By sorting the preoperative etiologies into two groups, perforation carries a higher risk of surgical mortality than bleeding.
1 November 2018 in The American Journal of Emergency Medicine
1 September 2020 in Risk Management and Healthcare Policy
Purpose: High prevalence of psychiatric comorbidities (PCs) has been widely documented in caustic substance ingestion cases. However, their effect on the clinical features and prognostic outcomes remains unclear due to the paucity of discussion. We report on detailed clinical courses with long-term multifaceted outcomes and review the association between caustic ingestion and each specific PC. Patients and Methods: The retrospective chart review included 396 adults (median follow-up, 16.6 months) with and 377 without (control group) PCs treated between 1999 and 2018 at Chang Gung Memorial Hospital. All PCs were diagnosed/confirmed by psychiatrists through face-to-face interviews. Results: The PCs predicted serious esophagogastroduodenoscopy grading, higher rates of admission/surgery/intensive care unit stay, increments of systemic/gastrointestinal complications, and poorer 5-year overall survival rates. The poor survival among patients with PCs was highly consistent with their baseline characteristics. Significantly advanced age, more non-PCs, alcoholism, illicit drug abuse, and baseline unhealthy status resulted in statistically higher risks of severe complications and limited recovery. Conclusion: PCs changed clinical patterns and had critical roles in the survival outcomes of caustic injury victims. Clinical awareness achieves benefit by limiting injuries in mild cases or allowing emergent interventions in severe cases. Future studies based on worldwide populations are essential for realizing geographic differences.
28 August 2020 in Journal of Medical Internet Research
Background: The application of mobile health (mHealth) platforms to monitor recovery in the postdischarge period has increased in recent years. Despite widespread enthusiasm for mHealth, few studies have evaluated the usability and user experience of mHealth in patients with surgical drainage. Objective: Our objectives were to (1) develop an image-based smartphone app, SurgCare, for postdrainage monitoring and (2) determine the feasibility and clinical value of the use of SurgCare by patients with drainage. Methods: We enrolled 80 patients with biliary or peritoneal drainage in this study. A total of 50 patients were assigned to the SurgCare group, who recorded drainage monitoring data with the smartphone app; and 30 patients who manually recorded the data were assigned to the conventional group. The patients continued to record data until drain removal. The primary aim was to validate feasibility for the user, which was defined as the proportion of patients using each element of the system. Moreover, the secondary aim was to evaluate the association of compliance with SurgCare and the occurrence of unexpected events. Results: The average submission duration was 14.98 days, and the overall daily submission rate was 84.2%. The average system usability scale was 83.7 (SD 3.5). This system met the definition of “definitely feasible” in 34 patients, “possibly feasible” in 10 patients, and “not feasible” in 3 patients. We found that the occurrence rates of complications in the SurgCare group and the conventional group were 6% and 26%, respectively, with statistically significant differences P=.03. The rate of unexpected hospital return was lower in the SurgCare group (6%) than in the conventional groups (26%) (P=.03). Conclusions: Patients can learn to use a smartphone app for postdischarge drainage monitoring with high levels of user satisfaction. We also identified a high degree of compliance with app-based drainage-recording design features, which is an aspect of mHealth that can improve surgical care.