Objectives: Platelet-rich plasma (PRP) is an autologous product derived from whole blood through the process of gradient density centrifugation. Autologous PRP has been shown to be safe and effective in promoting natural wound healing processes. This study assessed the efficacy of fat graft myringoplasty combined with PRP in tympanic membrane (TM) healing in dry TM perforation due to chronic otitis media and compared it to that of fat graft myringoplasty. Methods: A systematic review and meta-analysis of prospective or retrospective studies was conducted, with database searches (PubMed, Cochrane Library, Embase, Web of Science, Scopus, and Google Scholar) performed through April 2025. The outcomes included the success rate (complete closure of the TM postoperatively) and changes in air-bone conduction gap (ABG) (preoperatively and postoperatively) in the audiogram. Results: Five studies (n=522) were included. Fat myringoplasty with PRP resulted in greater improvement in ABGs (standardized mean difference [SMD], 0.329; 95% CI, 0.132 to 0.527) and higher success rates (odds ratio [OR], 3.129; 95% CI, 1.882 to 5.203) compared to fat myringoplasty alone. In medium-sized perforations, combination therapy demonstrated superior outcomes in both success rate (OR, 3.179; 95% CI, 1.891 to 5.343) and ABG (SMD, 0.357; 95% CI, 0.128 to 0.586). However, no significant differences were observed in cases involving small and medium-sized perforations (success rate: OR, 2.154; 95% CI, 0.174 to 26.672 and changes in ABG: SMD, 0.249; 95% CI, –0.141 to 0.638). Both liquid and gel PRP significantly improved ABGs and TM closure rates, with no significant difference between types. Conclusion This study demonstrated a higher success rate and greater hearing improvement with PRP-assisted fat myringoplasty, especially in medium-sized perforations. Both liquid and gel PRP forms were effective, with no significant difference between them.

This study aimed to develop and evaluate a multimodal deep learning model that integrates panoramic radiographs and clinical data to improve the diagnosis of medication-related osteonecrosis of the jaw (MRONJ). Materials and Methods:The study included 705 panoramic radiographs (401 MRONJ and 304 Normal) collected from Ewha Womans University Mokdong and Seoul Hospitals. Radiographs were paired with patient information, including age, sex, body mass index (BMI), type of medication, duration, route of administration, dosage, and dental risk factors. Two deep learning models were compared: a ResNet-50 model using panoramic radiographs alone and the Modified ResNet-50 model incorporating both image and clinical features. Model performance was evaluated using accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC). Lesion localization and feature importance were analyzed using Gradient-weighted Class Activation Mapping (Grad-CAM) and SHapley Additive exPlanations (SHAP). Result: Significant differences in clinical characteristics were observed between the groups; the MRONJ group was older (74.5±10.0 vs. 68.6±12.7 years, P<0.01) and had a higher proportion of females (92.5% vs. 82.2%, P<0.01). In model performance, the Modified ResNet-50 model outperformed the PR-only model. Specifically, the Modified ResNet-50 model achieved an accuracy of 75.0%, a recall of 84.3%, and an AUC of 0.843, whereas the PR-only model showed lower performance with an accuracy of 72.1%, a recall of 81.4%, and an AUC of 0.720. SHAP analysis identified prolonged medication use and specific drug types (e.g., zoledronate, P<0.01) as influential predictors. Conclusion: The proposed multimodal deep learning framework demonstrated improved diagnostic performance and interpretability compared to radiograph-only models. This approach may serve as a practical and explainable AI-assisted tool for MRONJ diagnosis in routine dental practice.

Purpose: This study explored the feasibility of utilizing a deep learning artificial intelligence (AI) model to detect ileocolic intussusception on grayscale ultrasound images. Methods: This retrospective observational study incorporated ultrasound images of children who underwent emergency ultrasonography for suspected ileocolic intussusception. After excluding video clips, Doppler images, and annotated images, 40,765 images from two tertiary hospitals were included (positive-to-negative ratio: hospital A, 2,775:35,373; hospital B, 140:2,477). Images from hospital A were split into a training set, a tuning set, and an internal test set (ITS) at a ratio of 7:1.5:1.5. Images from hospital B comprised an external test set (ETS). For each image indicating intussusception, two radiologists provided a bounding box as the ground-truth label. If intussusception was suspected in the input image, the model generated a bounding box with a confidence score (0-1) at the estimated lesion location. Average precision (AP) was used to evaluate overall model performance. The performance of practical thresholds for the modelgenerated confidence score, as determined from the ITS, was verified using the ETS. Results: The AP values for the ITS and ETS were 0.952 and 0.936, respectively. Two confidence thresholds, CTopt and CTprecision, were set at 0.557 and 0.790, respectively. For the ETS, the perimage precision and recall were 95.7% and 80.0% with CTopt, and 98.4% and 44.3% with CTprecision. For per-patient diagnosis, the sensitivity and specificity were 100.0% and 97.1% with CTopt, and 100.0% and 99.0% with CTprecision. The average number of false positives per patient was 0.04 with CTopt and 0.01 for CTprecision. Conclusion The feasibility of using an AI model to diagnose ileocolic intussusception on ultrasonography was demonstrated. However, further study involving bias-free data is warranted for robust clinical validation.

Purpose: This study explored the feasibility of utilizing a deep learning artificial intelligence (AI) model to detect ileocolic intussusception on grayscale ultrasound images. Methods: This retrospective observational study incorporated ultrasound images of children who underwent emergency ultrasonography for suspected ileocolic intussusception. After excluding video clips, Doppler images, and annotated images, 40,765 images from two tertiary hospitals were included (positive-to-negative ratio: hospital A, 2,775:35,373; hospital B, 140:2,477). Images from hospital A were split into a training set, a tuning set, and an internal test set (ITS) at a ratio of 7:1.5:1.5. Images from hospital B comprised an external test set (ETS). For each image indicating intussusception, two radiologists provided a bounding box as the ground-truth label. If intussusception was suspected in the input image, the model generated a bounding box with a confidence score (0-1) at the estimated lesion location. Average precision (AP) was used to evaluate overall model performance. The performance of practical thresholds for the modelgenerated confidence score, as determined from the ITS, was verified using the ETS. Results: The AP values for the ITS and ETS were 0.952 and 0.936, respectively. Two confidence thresholds, CTopt and CTprecision, were set at 0.557 and 0.790, respectively. For the ETS, the perimage precision and recall were 95.7% and 80.0% with CTopt, and 98.4% and 44.3% with CTprecision. For per-patient diagnosis, the sensitivity and specificity were 100.0% and 97.1% with CTopt, and 100.0% and 99.0% with CTprecision. The average number of false positives per patient was 0.04 with CTopt and 0.01 for CTprecision. Conclusion The feasibility of using an AI model to diagnose ileocolic intussusception on ultrasonography was demonstrated. However, further study involving bias-free data is warranted for robust clinical validation.

Purpose: This study explored the feasibility of utilizing a deep learning artificial intelligence (AI) model to detect ileocolic intussusception on grayscale ultrasound images. Methods: This retrospective observational study incorporated ultrasound images of children who underwent emergency ultrasonography for suspected ileocolic intussusception. After excluding video clips, Doppler images, and annotated images, 40,765 images from two tertiary hospitals were included (positive-to-negative ratio: hospital A, 2,775:35,373; hospital B, 140:2,477). Images from hospital A were split into a training set, a tuning set, and an internal test set (ITS) at a ratio of 7:1.5:1.5. Images from hospital B comprised an external test set (ETS). For each image indicating intussusception, two radiologists provided a bounding box as the ground-truth label. If intussusception was suspected in the input image, the model generated a bounding box with a confidence score (0-1) at the estimated lesion location. Average precision (AP) was used to evaluate overall model performance. The performance of practical thresholds for the modelgenerated confidence score, as determined from the ITS, was verified using the ETS. Results: The AP values for the ITS and ETS were 0.952 and 0.936, respectively. Two confidence thresholds, CTopt and CTprecision, were set at 0.557 and 0.790, respectively. For the ETS, the perimage precision and recall were 95.7% and 80.0% with CTopt, and 98.4% and 44.3% with CTprecision. For per-patient diagnosis, the sensitivity and specificity were 100.0% and 97.1% with CTopt, and 100.0% and 99.0% with CTprecision. The average number of false positives per patient was 0.04 with CTopt and 0.01 for CTprecision. Conclusion The feasibility of using an AI model to diagnose ileocolic intussusception on ultrasonography was demonstrated. However, further study involving bias-free data is warranted for robust clinical validation.

Purpose: This study explored the feasibility of utilizing a deep learning artificial intelligence (AI) model to detect ileocolic intussusception on grayscale ultrasound images. Methods: This retrospective observational study incorporated ultrasound images of children who underwent emergency ultrasonography for suspected ileocolic intussusception. After excluding video clips, Doppler images, and annotated images, 40,765 images from two tertiary hospitals were included (positive-to-negative ratio: hospital A, 2,775:35,373; hospital B, 140:2,477). Images from hospital A were split into a training set, a tuning set, and an internal test set (ITS) at a ratio of 7:1.5:1.5. Images from hospital B comprised an external test set (ETS). For each image indicating intussusception, two radiologists provided a bounding box as the ground-truth label. If intussusception was suspected in the input image, the model generated a bounding box with a confidence score (0-1) at the estimated lesion location. Average precision (AP) was used to evaluate overall model performance. The performance of practical thresholds for the modelgenerated confidence score, as determined from the ITS, was verified using the ETS. Results: The AP values for the ITS and ETS were 0.952 and 0.936, respectively. Two confidence thresholds, CTopt and CTprecision, were set at 0.557 and 0.790, respectively. For the ETS, the perimage precision and recall were 95.7% and 80.0% with CTopt, and 98.4% and 44.3% with CTprecision. For per-patient diagnosis, the sensitivity and specificity were 100.0% and 97.1% with CTopt, and 100.0% and 99.0% with CTprecision. The average number of false positives per patient was 0.04 with CTopt and 0.01 for CTprecision. Conclusion The feasibility of using an AI model to diagnose ileocolic intussusception on ultrasonography was demonstrated. However, further study involving bias-free data is warranted for robust clinical validation.

Purpose: This study explored the feasibility of utilizing a deep learning artificial intelligence (AI) model to detect ileocolic intussusception on grayscale ultrasound images. Methods: This retrospective observational study incorporated ultrasound images of children who underwent emergency ultrasonography for suspected ileocolic intussusception. After excluding video clips, Doppler images, and annotated images, 40,765 images from two tertiary hospitals were included (positive-to-negative ratio: hospital A, 2,775:35,373; hospital B, 140:2,477). Images from hospital A were split into a training set, a tuning set, and an internal test set (ITS) at a ratio of 7:1.5:1.5. Images from hospital B comprised an external test set (ETS). For each image indicating intussusception, two radiologists provided a bounding box as the ground-truth label. If intussusception was suspected in the input image, the model generated a bounding box with a confidence score (0-1) at the estimated lesion location. Average precision (AP) was used to evaluate overall model performance. The performance of practical thresholds for the modelgenerated confidence score, as determined from the ITS, was verified using the ETS. Results: The AP values for the ITS and ETS were 0.952 and 0.936, respectively. Two confidence thresholds, CTopt and CTprecision, were set at 0.557 and 0.790, respectively. For the ETS, the perimage precision and recall were 95.7% and 80.0% with CTopt, and 98.4% and 44.3% with CTprecision. For per-patient diagnosis, the sensitivity and specificity were 100.0% and 97.1% with CTopt, and 100.0% and 99.0% with CTprecision. The average number of false positives per patient was 0.04 with CTopt and 0.01 for CTprecision. Conclusion The feasibility of using an AI model to diagnose ileocolic intussusception on ultrasonography was demonstrated. However, further study involving bias-free data is warranted for robust clinical validation.

Purpose: This study aimed to compare tumor tissue DNA (ttDNA) and circulating tumor DNA (ctDNA) to explore the clinical applicability of ctDNA and to better understand clonal evolution in patients with metastatic colorectal cancer undergoing palliative first-line systemic therapy. Materials and Methods: We performed targeted sequencing analysis of 88 cancer-associated genes using germline DNA, ctDNA at baseline (baseline-ctDNA), and ctDNA at progressive disease (PD-ctDNA). The results were compared with ttDNA data. Results: Among 208 consecutively enrolled patients, we selected 84 (41 males; median age, 59 years; range, 35 to 90 years) with all four sample types available. A total of 202 driver mutations were found in 34 genes. ttDNA exhibited the highest mutation frequency (n=232), followed by baseline-ctDNA (n=155) and PD-ctDNA (n=117). Sequencing ctDNA alongside ttDNA revealed additional mutations in 40 patients (47.6%). PD-ctDNA detected 13 novel mutations in 10 patients (11.9%) compared to ttDNA and baseline-ctDNA. Notably, seven mutations in five patients (6.0%) were missense or nonsense mutations in APC, TP53, SMAD4, and CDH1 genes. In baseline-ctDNA, higher maximal variant allele frequency (VAF) values (p=0.010) and higher VAF values of APC (p=0.012), TP53 (p=0.012), and KRAS (p=0.005) mutations were significantly associated with worse overall survival. Conclusion While ttDNA remains more sensitive than ctDNA, our ctDNA platform demonstrated validity and potential value when ttDNA was unavailable. Post-treatment analysis of PD-ctDNA unveiled new pathogenic mutations, signifying cancer’s clonal evolution. Additionally, baseline-ctDNA’s VAF values were prognostic after treatment.

Viral load and the duration of viral shedding of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are important determinants of the transmission of coronavirus disease 2019.In this study, we examined the effects of viral doses on the lung and spleen of K18-hACE2 transgenic mice by temporal histological and transcriptional analyses. Approximately, 1×105 plaque-forming units (PFU) of SARS-CoV-2 induced strong host responses in the lungs from 2 days post inoculation (dpi) which did not recover until the mice died, whereas responses to the virus were obvious at 5 days, recovering to the basal state by 14 dpi at 1×102 PFU. Further, flow cytometry showed that number of CD8+ T cells continuously increased in 1×102 PFU-virusinfected lungs from 2 dpi, but not in 1×105 PFU-virus-infected lungs. In spleens, responses to the virus were prominent from 2 dpi, and number of B cells was significantly decreased at 1×105PFU; however, 1×102 PFU of virus induced very weak responses from 2 dpi which recovered by 10 dpi. Although the defense responses returned to normal and the mice survived, lung histology showed evidence of fibrosis, suggesting sequelae of SARS-CoV-2 infection. Our findings indicate that specific effectors of the immune response in the lung and spleen were either increased or depleted in response to doses of SARS-CoV-2. This study demonstrated that the response of local and systemic immune effectors to a viral infection varies with viral dose, which either exacerbates the severity of the infection or accelerates its elimination.

Background: A healthcare system’s collapse due to a pandemic, such as the coronavirus disease 2019 (COVID-19), can expose healthcare workers (HCWs) to various mental health problems. This study aimed to investigate the impact of the COVID-19 pandemic on the depression and anxiety of HCWs. Methods: A nationwide questionnaire-based survey was conducted on HCWs who worked in healthcare facilities and public health centers in Korea in December 2020. Patient Health Questionnaire-9 (PHQ-9) and Generalized Anxiety Disorder-7 (GAD-7) were used to measure depression and anxiety. To investigate factors associated with depression and anxiety, stepwise multiple logistic regression analysis was performed. Results: A total of 1,425 participating HCWs were included. The mean depression score (PHQ-9) of HCWs before and after COVID-19 increased from 2.37 to 5.39, and the mean anxiety score (GAD-7) increased from 1.41 to 3.41. The proportion of HCWs with moderate to severe depression (PHQ-9 ≥ 10) increased from 3.8% before COVID-19 to 19.5% after COVID-19, whereas that of HCWs with moderate to severe anxiety (GAD-7 ≥ 10) increased from 2.0% to 10.1%. In our study, insomnia, chronic fatigue symptoms and physical symptoms after COVID-19, anxiety score (GAD-7) after COVID-19, living alone, and exhaustion were positively correlated with depression. Furthermore, post-traumatic stress symptoms, stress score (Global Assessment of Recent Stress), depression score (PHQ-9) after COVID-19, and exhaustion were positively correlated with anxiety. Conclusion In Korea, during the COVID-19 pandemic, HCWs commonly suffered from mental health problems, including depression and anxiety. Regularly checking the physical and mental health problems of HCWs during the COVID-19 pandemic is crucial, and social support and strategy are needed to reduce the heavy workload and psychological distress of HCWs.