Background/Objectives: Acute ischemic stroke (AIS) is a leading cause of mortality and disability worldwide, with early and accurate diagnosis being critical for timely intervention and improved patient outcomes. This retrospective study aimed to assess the diagnostic performance of two advanced artificial intelligence (AI) models, Chat Generative Pre-trained Transformer (ChatGPT-4o) and Claude 3.5 Sonnet, in identifying AIS from diffusion-weighted imaging (DWI). Methods: The DWI images of a total of 110 cases (AIS group: n = 55, healthy controls: n = 55) were provided to the AI models via standardized prompts. The models’ responses were compared to radiologists’ gold-standard evaluations, and performance metrics such as sensitivity, specificity, and diagnostic accuracy were calculated. Results: Both models exhibited a high sensitivity for AIS detection (ChatGPT-4o: 100%, Claude 3.5 Sonnet: 94.5%). However, ChatGPT-4o demonstrated a significantly lower specificity (3.6%) compared to Claude 3.5 Sonnet (74.5%). The agreement with radiologists was poor for ChatGPT-4o (κ = 0.036; %95 CI: −0.013, 0.085) but good for Claude 3.5 Sonnet (κ = 0.691; %95 CI: 0.558, 0.824). In terms of the AIS hemispheric localization accuracy, Claude 3.5 Sonnet (67.2%) outperformed ChatGPT-4o (32.7%). Similarly, for specific AIS localization, Claude 3.5 Sonnet (30.9%) showed greater accuracy than ChatGPT-4o (7.3%), with these differences being statistically significant (p < 0.05). Conclusions: This study highlights the superior diagnostic performance of Claude 3.5 Sonnet compared to ChatGPT-4o in identifying AIS from DWI. Despite its advantages, both models demonstrated notable limitations in accuracy, emphasizing the need for further development before achieving full clinical applicability. These findings underline the potential of AI tools in radiological diagnostics while acknowledging their current limitations.

Background: This study aimed to investigate the effect of manual therapy on pain, kinesiophobia, pain catastrophizing, anxiety, depression, and quality of life in patients with lumbar disc herniation (LDH). Methods: The study included 32 LDH patients. Patients were divided into the Manual therapy group (MTG—age 39.81 ± 9.45 years) and the Exercise group (EG—age 38.31 ± 9.21 years) by sealed envelope randomization. Patients were evaluated pre-study, post-study, and after a 3-month period using the McGill–Melzack Pain Questionnaire (MMPQ), Hospital Anxiety and Depression Scale (HADS), Tampa Kinesiophobia Scale (TKS), Pain Catastrophizing Scale (PCS) and Nottingham Health Profile (NSP). The exercise group received a total of ten sessions of stabilization exercises and sham spinal mobilization in five weeks, two sessions per week. In addition to the stabilization exercises, mobilization applications including Anterior-Posterior Lumbar Spinal Mobilization, Lumbar Spinal Rotational Mobilization, and Joint Mobilization in Lumbar Flexion Position, were applied to the manual therapy group. Results: It was found that the HADS and TKS values decreased in the MTG group compared to the pre-treatment period (p < 0.05), while there was no difference between these values in the EG group (p > 0.05). There was a statistically significant difference in the MMPQ, PCS, and NHP values after treatment in both the MTG and EG groups (p < 0.05). Conclusions: It was found that manual therapy had positive effects on psychological factors such as pain, kinesiophobia, pain catastrophizing, anxiety, depression, and quality of life in patients with LDH. Trial registration: NCT05804357 (27 March 2023) (retrospectively registered).

Background/Objectives: Acute ischemic stroke (AIS) is a leading cause of mortality and disability worldwide, with early and accurate diagnosis being critical for timely intervention and improved patient outcomes. This retrospective study aimed to assess the diagnostic performance of two advanced artificial intelligence (AI) models, Chat Generative Pre-trained Transformer (ChatGPT-4o) and Claude 3.5 Sonnet, in identifying AIS from diffusion-weighted imaging (DWI). Methods: The DWI images of a total of 110 cases (AIS group: n=55, healthy controls: n=55) were provided to the AI models via standardized prompts. Their responses were compared to the gold-standard evaluations by radiologists, and performance metrics, including sensitivity, specificity, positive predictive value, negative predictive value, diagnostic accuracy, and inter-model agreement levels, were calculated. Results: Both models exhibited high sensitivity for AIS detection (ChatGPT-4o: 100%, Claude 3.5 Sonnet: 94.5%). However, ChatGPT-4o demonstrated significantly lower specificity (3.6%) compared to Claude 3.5 Sonnet (74.5%). Agreement with radiologists was poor for ChatGPT-4o (κ=0.036) but good for Claude 3.5 Sonnet (κ=0.691). In terms of hemispheric localization accuracy, Claude 3.5 Sonnet (67.2%) outperformed ChatGPT-4o (32.7%). Similarly, for specific AIS localization, Claude 3.5 Sonnet (30.9%) showed greater accuracy than ChatGPT-4o (7.2%), with these differences being statistically significant (p<0.05). Conclusions: This study highlights the superior diagnostic performance of Claude 3.5 Sonnet compared to ChatGPT-4o in identifying AIS from DWI. Despite its advantages, both models demonstrated notable limitations in accuracy, emphasizing the need for further development before achieving full clinical applicability. These findings underline the potential of AI tools in radiological diagnostics while acknowledging their current limitations.

Objective: Sarcopenia, characterized by progressive skeletal muscle loss, poses significant health risks, including physical impairment and mortality. The relationship between sarcopenia and insulin resistance suggests insulin therapy’s potential in preserving muscle mass, particularly in Type 2 diabetes mellitus (T2DM) patients. This study aims to evaluate the Psoas Muscle Index (PMI) via computed tomography (CT) in middle-aged T2DM patients on insulin therapy versus oral antidiabetic drugs (OAD) and controls. Methods: This retrospective study included 107 middle-aged T2DM patients undergoing non-contrast CT scans and 58 age-matched controls. CT images were analyzed to calculate PMI. Statistical analysis included Chi-square tests, independent samples t-tests, Mann–Whitney U tests, and correlation analyses. Results: Insulin-treated patients exhibited higher PMI than OAD users (p < 0.001), while OAD users had lower PMI than controls (p < 0.001). No significant difference was found between insulin-treated patients and controls (p = 0.616). Negative correlations were observed between T2DM duration/age and PMI across all groups, with a positive correlation between T2DM duration and BMI observed in the OAD group. Conclusions: Insulin therapy in T2DM patients, regardless of age or disease duration, positively impacts muscle mass, highlighting its potential in preserving muscular health and advocating for tailored treatment strategies in T2DM management.

In this study, a deep learning method for kidney stone detection is proposed. The method utilizes transfer learning by extracting features from a pre-trained ImageNet model. However, unlike traditional transfer learning, which directly applies or fine-tunes a pre-trained model, the proposed approach integrates a custom-designed CNN that operates in parallel with the pre-trained network. The feature maps obtained from both networks are fused to enhance the model's representation power. After this integration, task-specific classification layers are added, and the training process is conducted on both the classification layers and the optimized model. This approach improves the overall performance of the model while providing a more efficient training process. As part of this study, a new dataset was created, consisting of 2166 axial slice images from 241 patients and 2018 axial slice images from 46 healthy individuals. Experiments conducted using EfficientNetV2s, MobileNetV4s, SqueezeNet, and ResNet18-based models revealed that the EfficientNetV2s and MobileNetV4s-based models excelled in terms of accuracy, while the SqueezeNet and ResNet18-based models provided stronger results in terms of interpretability.

Background: Arteriovenous fistula (AVF) thrombosis remains a critical complication in hemodialysis (HD) patients, often leading to treatment delays and requiring urgent intervention. While endovascular therapy (EVT) is commonly employed, less invasive strategies such as percutaneous thrombolytic therapy are gaining attention due to their potential to restore patency and avoid more complex procedures. This study assessed the effectiveness of percutaneous thrombolytic therapy in acute AVF thrombosis and explored key predictors associated with the need for subsequent endovascular intervention. Methods: This retrospective study included 42 patients who underwent ultrasound-guided percutaneous thrombolytic therapy using low-dose alteplase (3–5 mg). Technical and clinical success, complication rates, and the need for additional EVT were assessed. Statistical analyses including logistic regression and ROC analysis were used to determine independent predictors for EVT. Results: The clinical success rate was 97.6%, with 69% of patients achieving AVF patency without EVT. Symptom duration emerged as the strongest predictor for EVT; patients with symptoms > 2.5 days had significantly higher EVT rates (p = 0.01). Each additional day of symptoms increased the odds of requiring EVT by 88.5% (OR = 1.885, p = 0.012). Female patients were also more likely to require EVT than males (p = 0.005). No significant associations were found for age, BMI, or fistula characteristics. Conclusion: Percutaneous thrombolytic therapy is a highly effective and minimally invasive option for acute AVF thrombosis. Symptom duration > 2.5 days is a key threshold predicting the need for EVT, highlighting the critical importance of early intervention. These findings may inform clinical decision-making and optimize access salvage strategies in dialysis patients.