RESUMEN

Background Multiple commercial artificial intelligence (AI) products exist for assessing radiographs; however, comparable performance data for these algorithms are limited. Purpose To perform an independent, stand-alone validation of commercially available AI products for bone age prediction based on hand radiographs and lung nodule detection on chest radiographs. Materials and Methods This retrospective study was carried out as part of Project AIR. Nine of 17 eligible AI products were validated on data from seven Dutch hospitals. For bone age prediction, the root mean square error (RMSE) and Pearson correlation coefficient were computed. The reference standard was set by three to five expert readers. For lung nodule detection, the area under the receiver operating characteristic curve (AUC) was computed. The reference standard was set by a chest radiologist based on CT. Randomized subsets of hand (n = 95) and chest (n = 140) radiographs were read by 14 and 17 human readers, respectively, with varying experience. Results Two bone age prediction algorithms were tested on hand radiographs (from January 2017 to January 2022) in 326 patients (mean age, 10 years ± 4 [SD]; 173 female patients) and correlated strongly with the reference standard (r = 0.99; P < .001 for both). No difference in RMSE was observed between algorithms (0.63 years [95% CI: 0.58, 0.69] and 0.57 years [95% CI: 0.52, 0.61]) and readers (0.68 years [95% CI: 0.64, 0.73]). Seven lung nodule detection algorithms were validated on chest radiographs (from January 2012 to May 2022) in 386 patients (mean age, 64 years ± 11; 223 male patients). Compared with readers (mean AUC, 0.81 [95% CI: 0.77, 0.85]), four algorithms performed better (AUC range, 0.86-0.93; P value range, <.001 to .04). Conclusions Compared with human readers, four AI algorithms for detecting lung nodules on chest radiographs showed improved performance, whereas the remaining algorithms tested showed no evidence of a difference in performance. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Omoumi and Richiardi in this issue.

Asunto(s)

RESUMEN

Chronic wounds are a significant burden on healthcare systems due to high costs of care (2%-4% total healthcare cost) and a considerable burden on patient's quality of life. Patient-reported outcome measures (PROMs) are questionnaires developed to enable patient self-assessments of their outcomes. A gap in knowledge exists because previous reviews on wound-specific PROMs did not evaluate the quality of the development. The main question is which PROM has the best quality development properties and should be used in clinical care and research. Methods: PubMed, Embase, and CINAHL were searched from their inception through December 2021. Studies that included patients aged 18 years or older, with chronic wounds, and who reported using a condition-specific PROM for wounds were extracted. We excluded generic PROMs, comments, guidelines, and editorial letters. The COSMIN-guidelines were used to evaluate the quality of the PROMs. Results: Of the 16,356 articles, a total of 251 articles describing 33 condition-specific PROMs for wounds were used. In total, 17 of 33 (52%) PROMs were developed for specific wound types, and nine of 33 (27%) PROMs were developed for any type of wound. Two of 33 (6%) PROMs were not rated because no development article was available. Only the SCI-QOL (Spinal Cord Injury-QOL) and the WOUND-Q rated "very good" in PROM design. Conclusions: Thirty-three condition-specific PROMs were found. Only the SCI-QOL and the WOUND-Q rated very good in PROM design. The WOUND-Q is the only condition-specific PROM, which can be used in all types of chronic wounds in any anatomic location.

RESUMEN

BACKGROUND: During the COVID-19 pandemic, a decrease in the number of patients presenting with acute appendicitis was observed. It is unclear whether this caused a shift towards more complicated cases of acute appendicitis. We compared a cohort of patients diagnosed with acute appendicitis during the 2020 COVID-19 pandemic with a 2019 control cohort. METHODS: We retrospectively included consecutive adult patients in 21 hospitals presenting with acute appendicitis in a COVID-19 pandemic cohort (March 15 - April 30, 2020) and a control cohort (March 15 - April 30, 2019). Primary outcome was the proportion of complicated appendicitis. Secondary outcomes included prehospital delay, appendicitis severity, and postoperative complication rates. RESULTS: The COVID-19 pandemic cohort comprised 607 patients vs. 642 patients in the control cohort. During the COVID-19 pandemic, a higher proportion of complicated appendicitis was seen (46.9% vs. 38.5%; p = 0.003). More patients had symptoms exceeding 24 h (61.1% vs. 56.2%, respectively, p = 0.048). After correction for prehospital delay, presentation during the first wave of the COVID-19 pandemic was still associated with a higher rate of complicated appendicitis. Patients presenting > 24 h after onset of symptoms during the COVID-19 pandemic were older (median 45 vs. 37 years; p = 0.001) and had more postoperative complications (15.3% vs. 6.7%; p = 0.002). CONCLUSIONS: Although the incidence of acute appendicitis was slightly lower during the first wave of the 2020 COVID-19 pandemic, more patients presented with a delay and with complicated appendicitis than in a corresponding period in 2019. Spontaneous resolution of mild appendicitis may have contributed to the increased proportion of patients with complicated appendicitis. Late presenting patients were older and experienced more postoperative complications compared to the control cohort.

Asunto(s)