SARS-CoV-2 damages not only the lungs, but also the liver and kidney. Most critically ill COVID-19 patients have liver and kidney dysfunctions. The early identification of patients with COVID-19 who will develop severe or critical disease symptoms is important for delivering proper and early treatment. This research studies the correlation of liver and kidney function indexes and COVID-19 outcomes. Electronic medical record data from 391 patients diagnosed with COVID-19 in the COVID-19 Department, Galilee Medical Center, Nahariya, Israel were collected. Epidemiological, clinical, laboratory, and imaging variables were analyzed. The liver and kidney enzyme indexes were measured upon admission and discharge. A correlation between laboratory levels and severity and mortality of COVID-19 patients was undertaken. This study included 391 COVID-19 patients, 258 mild patients and 133 severe patients. Multivariate stepwise regression analyses and discriminant analyses were used to identify and validate powerful predictors. The main outcome was death or invasive ventilation. Three factors, namely higher urea nitrogen (BUN) and IL-6, and lower albumin levels, were the most powerful predictors of mortality, and classified the results (survival vs. death) correctly in 85% of cases (diagnostic accuracy) with a sensitivity of 88% and a specificity 55%. Compared with mild patients, severe patients had lower albumin (ALB), higher alanine aminotransferase (ALT), aspartate aminotransferase (AST) and BUN (all p < 0.001). COVID-19 patients, especially severe patients, have damage to liver and kidney function. BUN, IL-6 and albumin are factors predicting mortality while fibrinogen and AST could be independent factors for predicting the severity of COVID-19.

Aims: Extremity sarcomas (ES) are rare, aggressive malignancies requiring complex surgical decisions. While limb-salvage surgery (LSS) is the preferred treatment to preserve functionality, local disease progression can necessitate secondary amputation. The oncological outcomes of secondary amputation compared to primary amputation remain uncertain, particularly with long-term follow-up. This study aimed to compare overall survival (OS), metastasis-free survival (MFS), local recurrence-free survival (LRFS), and postoperative complications between ES patients undergoing primary amputation (V1) and those requiring secondary amputation after failing LSS (V2), with a minimum follow-up of five years. Methods: A retrospective review was conducted at a national sarcoma center, including 71 patients treated between 2007 and 2017. Patients were categorized into V1 (n = 28) and V2 (n = 43) groups. Clinical and oncological data were collected from medical records and imaging, including tumor stage, surgical margins, and postoperative complications. All patients were followed up for a minimum of five years or until death. Kaplan–Meier survival analysis was performed to evaluate OS, MFS, and LRFS. Results: OS was 25% in the V1 group and 39.5% in the V2 group (p = 0.6). MFS (10.5 months, p = 0.2) and LRFS (27.4 vs. 34.4 months, p = 0.6) were comparable between groups. Postoperative complications occurred in 34.9% (V1) and 32.1% (V2) of patients, with infections being the most common complication. Patients with complications exhibited shorter MFS (p = 0.029). Negative surgical margins were achieved at 96.4% (V1) and 97.6% (V2). Conclusions: Secondary amputation following failing LSS demonstrates similar oncological outcomes to primary amputation, even with a minimum follow-up of five years. These findings support LSS as the preferred initial approach for ES patients. Postoperative complications associated with reduced MFS underscore the need for rigorous postoperative protocols. A multidisciplinary approach remains essential for optimizing long-term outcomes.

Background: Dedicated artificial intelligence (AI) systems for fracture detection already exist, yet general-purpose multimodal models are increasingly accessible to clinicians despite not being developed or formally validated as medical devices. Their behavior in focused orthopedic imaging tasks remains insufficiently characterized. Purpose: To characterize how two accessible general-purpose multimodal models interpret AP pelvis radiographs with hip fractures, focusing on context dependence, overconfidence, and complementary error patterns within a surgically confirmed positive-only cohort. This was a behavioral characterization study of a fracture-positive cohort, not a diagnostic accuracy evaluation. Methods: In April 2026, we retrospectively studied 214 surgically confirmed hip fractures on AP pelvis radiographs using two general-purpose multimodal models under six prompting conditions. In runs A–D, the models were explicitly told that a hip fracture was present and were asked to classify it; in runs E–F, they were not told whether a hip fracture was present. Each image was rerun de novo in a separate chat session through vendor APIs using a fixed base prompt and no image preprocessing. We recorded hip-fracture detection, correct laterality, coarse fracture pattern, intracapsular displacement, AO/OTA grading, subtrochanteric identification, and self-reported confidence. Because the cohort contained hip fractures only, we report fracture-detection rates and classification performance within a positive-only cohort rather than full diagnostic-accuracy metrics. Results: Using the more conservative endpoint of hip-fracture detection with correct laterality, GPT-5.4 was correct in 79.0% and 86.4% of cases in runs E and F, whereas Gemini was correct in 80.4% and 93.5%, respectively. When outputs from both models were combined, this endpoint reached 89.7% in run E and 96.7% in run F, indicating complementary rather than redundant error patterns. Incorrect laterality cues markedly degraded performance, from 90.7% to 66.4% in GPT-5.4 and from 97.7% to 57.0% in Gemini. Performance remained limited for treatment-relevant subtyping, particularly AO/OTA grading and subtrochanteric identification. Both models frequently remained highly confident when wrong, and self-reported confidence did not reliably distinguish correct from incorrect outputs. Conclusions: Accessible general-purpose multimodal models showed partial capability for coarse hip-fracture interpretation, but they remained context-sensitive, unreliable for treatment-relevant subtyping, and highly confident even when incorrect. Their complementary error patterns are hypothesis-generating rather than evidence of clinical readiness. On the basis of these findings, we do not support unvalidated or uncontrolled clinical use of such models. As access to these tools expands, explicit usage boundaries, minimum performance expectations, repeated local revalidation, and sustained human oversight become increasingly necessary.

Osteoid osteoma often requires surgical intervention following conservative treatment failure. While traditional radiofrequency ablation (RFA) is effective, poor prognostic factors such as tumor size, anatomical location, patient age, sex, and previous interventions may negatively influence outcomes. Cone Beam Computed Tomography (CBCT)-navigated RFA could mitigate these negative predictors by enhancing procedural accuracy; however, its specific impact on these factors remains under-investigated. This retrospective study included 127 patients (median age 20, IQR: 16-25) undergoing CBCT-navigated RFA for osteoid osteomas from 2015 to 2021. Outcomes assessed were treatment failure, recurrence rates, pain relief (Visual Analog Scale [VAS]), functional improvement (Eastern Cooperative Oncology Group [ECOG] status), and postoperative complications classified by the Society of Interventional Radiology (SIR) criteria. Multivariate logistic regression and statistical analyses were conducted to evaluate the influence of previously identified poor prognostic factors on outcomes. Significant long-term pain relief occurred in 97 % of patients, with median VAS scores improving from 9 to 0 postoperatively (p < 0.01). Median ECOG scores improved significantly from 1 to 0 (p < 0.01). Overall treatment failure was observed in 11(8.7 %) patients, including symptomatic recurrence (4, 3.1 %), minor peripheral nerve complaints (3, 2.36 %), limited range of motion (3, 2.36 %), and Metatarsal Fracture (1, 0.8 %). Adult age was the only factor significantly associated with treatment failure (OR = 1.117, p = 0.036), whereas tumor size, anatomical location, biological sex, and prior interventions showed no significant associations. CBCT-navigated RFA effectively reduces the negative impact of traditionally poor prognostic factors, achieving high success rates with minimal complications. While older age remains a modest predictor of poorer outcomes, the absence of significant associations with other previously identified risk factors underscores the enhanced precision and clinical value of CBCT navigation.

Introduction: Cut-out is the most consequential mechanical complication after proximal femoral nailing and requires prompt recognition. General-purpose artificial intelligence (AI) models with image-interpretation capability are increasingly accessible, yet their diagnostic performance for this task is unknown. We evaluated ChatGPT’s accuracy for predicting cut-out following proximal femoral nailing and compared TFNA and Gamma nail subgroups. Materials and methods: In this retrospective predictive-accuracy study, 989 patients (683 TFNA, 306 Gamma nail; mean age 83.8 years; cut-out prevalence 2.5%) were analysed. For each case, three baseline radiographs — the injury anteroposterior and lateral views and the immediate postoperative control radiograph, all obtained before any complication could be radiographically present — together with the full clinical data set (excluding complication-related information) were submitted to ChatGPT, which returned a binary prediction (yes/no) and probability estimate (0–100%). The follow-up radiographs on which cut-out becomes apparent were not provided to the model. Ground truth was established on subsequent follow-up radiographs by a departmental review board requiring agreement of two senior surgeons. Accuracy metrics were calculated with Wilson 95% confidence intervals (CI); exact McNemar’s and Mann-Whitney U tests assessed directional bias and calibration. Reporting followed STARD guidelines. Results: ChatGPT showed a sensitivity of 68.0% (95% CI 48.4–82.8%), specificity of 62.7% (59.6–65.7%), PPV of 4.5% (2.8–7.1%), NPV of 98.7% (97.4–99.3%), and AUC of 0.694 (0.580–0.790). Performance was higher for TFNA (sensitivity 73.7%, AUC 0.730) than for Gamma nail (sensitivity 50.0%, AUC 0.606). The model over-predicted cut-out, with 360 false positives against 8 false negatives (45:1; McNemar’s p < 0.001, Holm-corrected). Calibration was inverted: the median predicted probability was lower for correct (9%) than for incorrect (40%) classifications (Mann-Whitney p < 0.001). Conclusions: ChatGPT demonstrated limited, clinically insufficient accuracy for predicting cut-out following proximal femoral nailing, with a prohibitive false-positive burden and inverted calibration. Subgroup point estimates were lower for Gamma nail cases, but this exploratory difference did not reach statistical significance. General-purpose AI models are not currently suitable as a substitute for, or adjunct to, surgeon-led surveillance of this complication. Task-specific training, external validation, and defined clinical boundaries are required before such tools can be considered for fracture follow-up pathways.