Background: Depth of invasion (DOI) in oral tongue squamous cell carcinoma (OTSCC) is an important prognostic factor influencing tumor staging, surgical planning, and overall prognosis. Objectives: This study aimed to assess the correlation between computed tomography-based radiologic DOI (rDOI) and pathologic DOI (pDOI) and to evaluate the diagnostic accuracy of computed tomography-based rDOI, using pDOI as the gold standard. Patients and Methods: This retrospective, single-center diagnostic accuracy study included 60 consecutive eligible patients with biopsy-proven OTSCC who underwent preoperative contrast-enhanced computed tomography at an academic referral center between April 2022 and March 2023. Patients with recurrent tumors, prior head-and-neck radiotherapy, computed tomography artifacts precluding measurement, or missing pathological DOI were excluded. The index test, computed tomography-derived rDOI, was measured independently by two blinded radiologists with 5 and 20 years of head-and-neck imaging experience and compared with histopathological DOI as the reference standard, assessed by a dedicated head-and-neck pathologist blinded to imaging findings. The primary analysis evaluated agreement and diagnostic accuracy. Inter-reader reliability and the association between DOI and nodal involvement were prespecified secondary objectives. Results: Sixty patients were analyzed, including 35 men and 25 women, with a mean age of 54.0 ± 14.2 years. The mean pDOI was 10.71 ± 6.47 mm. Computed tomography systematically overestimated the pDOI by 1.12 mm for Radiologist 1 and by 2.19 mm for Radiologist 2. Lymph node involvement was present in 15 patients (25%), who had a significantly higher pDOI than those without lymph node involvement (14.13 vs. 8.91 mm; P = 0.002). Strong agreement was observed between pDOI and rDOI, particularly for DOI > 5 mm. For Radiologist 1, κ = 0.797 (95% confidence interval [CI]: 0.69 - 0.91), Spearman's rho = 0.668 (95% CI: 0.53 - 0.77), and P < 0.001. For Radiologist 2, κ = 0.700 (95% CI: 0.59 - 0.81) and P < 0.001. For detecting DOI > 5 mm, computed tomography showed an area under the curve of 0.934 for Radiologist 1 and 0.921 for Radiologist 2, with sensitivity > 89% and specificity > 85%. Inter-reader agreement was excellent (κ = 0.838; 95% CI: 0.76 - 0.92; Spearman's rho = 0.945; P < 0.001) and was independent of patient sex or lymph node status. Conclusion: Computed tomography-based rDOI is a reliable and diagnostically accurate preoperative tool for OTSCC staging, particularly for tumors with DOI > 5 mm. It may serve as a useful preoperative surrogate estimate of DOI when surgical pathology is not yet available. Ultrasound may be more accurate for superficial lesions with DOI ≤ 5 mm.
Context: Artificial intelligence (AI) has a critical role in hospitals that have successfully implemented it and achieved international rankings, because these institutions can serve as benchmarks for other hospitals. The main objective of this article was to examine the characteristics of pioneer hospitals in AI, with a focus on radiology departments. Evidence Acquisition: This narrative review was performed according to SANRA guidelines. Data were collected by searching PubMed, Google Scholar, and ScienceDirect for articles published between January 2019 and June 2025, using search strings that included ('artificial intelligence' OR 'AI') AND ('pioneer hospital') AND ('adoption' OR 'implementation') AND ('radiology'). After screening and selecting 41 papers for the final study, the characteristics of pioneer hospitals were determined. Two independent reviewers then confirmed the common features of pioneer hospitals through discussion and consensus. Risk of bias and reliability were assessed based on study design and transparency of reporting, and grey literature was used cautiously and clearly marked with access dates. Results: Hospital type, innovation, transparency, strong leadership, learning, and hospital size were identified as characteristics of pioneer hospitals. These hospitals have used AI in radiology departments to analyze data, predict outcomes, support clinical decision-making and operational efficiency, and improve diagnostic capabilities and workflow. Conclusion: Successful AI integration in pioneer radiology departments depends on synergy between technical readiness and organizational factors. Hospitals can enhance diagnostic accuracy and operational efficiency by prioritizing strong leadership, a collaborative culture, and a clear digital strategy alongside technological investment.
Context: Deep learning reconstruction is increasingly proposed to accelerate prostate T2-weighted magnetic resonance imaging (MRI) acquisition without loss of diagnostic quality. However, the magnitude of image quality benefits and the influence of deep learning network design remain unclear. Objectives: We aimed to compare the image quality of deep learning versus conventional prostate T2 MRI reconstruction and assess how acceleration factors and network architecture influence performance. Methods: A PRISMA search of PubMed/Medline, Embase, Web of Science, and Scopus (22 February 2026) identified studies comparing deep learning with conventional T2 MRI reconstruction. Paired standardized mean change (SMCC) values were pooled with a three-level random-effects model from extracted paired image-quality scores; meta-regressions examined scan-time acceleration. Small-study bias was tested with a precision-effect method. Results: Five studies (602 participants; 20 comparisons) included in the study. Overall deep learning reconstruction benefit was +0.14 standard deviation (SD) [95% confidence interval (CI) -2.00 to 2.28; I2 = 99 %]. At equal scan times deep learning reconstruction improved quality by +2.09 SD; for each fold increase in acceleration, image quality declined by 0.48 SD (95% CI -0.91 to -0.05; P = 0.030). C-SENSE AI at acceleration factors of 1.7, 3.4, and 4.8 demonstrated superior performance, whereas CycleGAN showed inferior results. Small-study effects were evident. Conclusion: Deep learning reconstruction can increase prostate T2-weighted MRI speed without significant changes in overall image quality; however, overly aggressive implementations may lead to image degradation. Careful validation and protocol-specific tuning are therefore essential prior to clinical adoption.
Introduction: Spontaneous ureteropelvic rupture (SUPR) is rare, and its combination with ureteral intraluminal dissection (UID) is even rarer. Case Presentation: A 62-year-old Chinese man was admitted to the hospital with acute, severe right lumbar pain accompanied by nausea and vomiting. Contrast-enhanced computed tomography of the abdomen and pelvis showed a double-lumen appearance and septum formation in the middle and upper segments of the right ureter during the excretory phase. Based on these imaging findings, UID caused by SUPR was diagnosed. The patient was managed conservatively with antibiotics for three weeks without ureteral stenting. Conclusion: This case involved SUPR combined with UID; however, the source remained elusive. Limited clinical awareness and variable clinical manifestations often make the diagnosis challenging. Diagnosis relies on delayed contrast-enhanced computed tomography (CT) of the abdomen and pelvis. Currently, no standardized guidelines are available, and minimally invasive endourological approaches are generally accepted as the preferred first-line option.
Background: Prostate carcinoma is the third most common malignancy among Malaysian men, and the skeleton is the most frequent metastatic site. Differentiating benign enostoses from osteoblastic metastases on computed tomography (CT) remains challenging. Although 99mTc-MDP bone scintigraphy is the gold standard, capacity constraints often delay treatment. Objectives: To evaluate the diagnostic performance of CT Hounsfield Unit (HU) measurements in differentiating these lesions and their correlation with systemic biological markers. Patients and Methods: We retrospectively evaluated 1041 sclerotic lesions (860 metastases and 181 benign enostoses) from 105 patients with prostate carcinoma. Mean HU values were recorded. Diagnostic metrics were computed at the lesion level, with adjustment for within-patient clustering using Generalized Estimating Equations (GEE). Optimal thresholds were determined using Receiver Operating Characteristic (ROC) analysis. Biological correlations with Gleason scores and Prostate-Specific Antigen (PSA) levels were analyzed using Spearman correlation and validated with GEE. Results: Mean HU demonstrated exceptional discriminatory power (area under the curve [AUC] = 0.984; 95% confidence interval [CI]: 0.975 - 0.993; P < 0.001). The optimal 944.99 HU threshold yielded 90.6% sensitivity, 97.8% specificity, and 99.5% positive predictive value (PPV). After GEE adjustment, histological Gleason grade (P = 0.098), systemic PSA (P = 0.301), and regional density differences (P > 0.05) showed no significant association with lesion density. Conclusion: Quantitative CT attenuation is a highly accurate triage adjunct. In a high-pretest-probability oncological setting, the 944.99 HU threshold confidently rules in benign enostoses, allowing clinicians to safely avoid unnecessary biopsies and optimize nuclear imaging resources.
Background: Breast cancer molecular subtyping is crucial for prognosis and treatment planning, prompting interest in noninvasive ultrasound-based approaches for identifying relevant correlations. Objectives: Breast cancer, which is highly prevalent worldwide, comprises several molecular subtypes, including luminal A (LA), luminal B (LB), triple-negative (TN), and human epidermal growth factor receptor 2 positive (HER2+). These subtypes have distinct prognostic and therapeutic implications. However, differences in ultrasound features among these subtypes and their potential associations with molecular classification remain underexplored. Methods: This retrospective cross-sectional study analyzed data from 140 women with primary invasive breast cancer treated at a referral hospital between April 2022 and March 2023. Standard ultrasound imaging and comprehensive clinicopathological data, including molecular subtypes identified by immunohistochemistry, were evaluated. Statistical analyses were performed using SPSS version 22 to assess correlations between ultrasound features and molecular subtypes. Results: The mean age of the participants was 49.75 years, and the mean tumor size was 27.1 mm. LA was the most common subtype (48%), followed by LB (25%), TN (18%), and HER2+ (7%). Adjusted associations between ultrasound features and molecular subtypes showed a nonsignificant trend, particularly for calcification, tumor shape, and tumor location. The TN subtype had the highest calcification rate, followed by the LA subtype (adjusted P = 0.107 ) Irregular shapes were common across all subtypes, whereas variation in the frequencies of round and oval shape & aogon; suggested potential subtype specific differences. Conclusions: Certain ultrasound features showed nonsignificant trends across molecular subtypes; however, these findings do not currently support reliable differentiation among subtypes. Further research is needed to validate these preliminary results and potentially expand the diagnostic role of ultrasound in breast cancer subtyping.
Context: Imaging is crucial in evaluating women with suspected appendiceal endometriosis (AE), as the condition often mimics acute or chronic appendicitis and presents a diagnostic challenge. While modalities like ultrasound (US), computed tomography (CT), and magnetic resonance imaging (MRI) can help identify abnormalities, their findings are frequently nonspecific. Therefore, awareness of imaging features is essential for accurate diagnosis and management, though definitive confirmation still relies on histopathological examination after surgical excision. Objectives: The present study aimed to review and investigate imaging findings in symptomatic AE. Methods: This systematic review was performed according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines. PubMed, Scopus, Web of Science, CINAHL Plus, and the Cochrane Library were searched using keywords including appendix, endometriosis, MRI, transvaginal sonography (TVS), and transrectal high intensity focused US. Studies were included if they reported imaging findings in symptomatic AE. Exclusion criteria were randomized controlled trials, controlled case studies, review articles, cohort studies, systematic reviews, conference abstracts, articles without full text, and non-English language articles. Study selection and data extraction were performed independently by two reviewers. The quality of included case reports was independently assessed using the Joanna Briggs Institute (JBI) critical appraisal checklist for case reports. Results: Twenty-six out of the total number of patients who underwent CT (30) had positive findings (86.6%), while 7 out of the total number of patients who underwent MRI (11) and 11 out of the total number of patients who underwent sonography (13) also had positive findings (63.6% and 84.6%, respectively). The mean age of the patients was 37.2 ± 7.07 years. Out of the total sample, 8 patients were pregnant. The overall imaging findings were: Normal (6 cases, 15%), wall thickening (9 cases, 22.5%), mass (15 cases, 37.5%), cystic mass (1 case, 2.5%), solid lesion in the left ovary (1 case, 2.5%), mucocele (3 cases, 7.5%), intussusception (4 cases, 10%), obstruction (5 cases, 12.5%), suspected obstruction (1 case, 2.5%), appendicitis (4 cases, 10%), fluid (11 cases, 27.5%), and abscess (3 cases, 7.5%). Conclusion: Right lower quadrant (RLQ) mass and bowel wall thickening are the most commonly reported findings in patients with AE. Further studies are required to retrospectively evaluate the imaging findings of the appendix in pathologically confirmed AE after pelvic surgery.
Background: Quantitative assessment of spleen volume on CT is clinically relevant but is often approximated using linear measurements. Deep learning-based segmentation enables automated volumetry; however, population-specific normative spleen volume models for Chinese adults remain limited. Objectives: To develop a deep learning-based automated CT segmentation model for normal spleens and to derive a Chinese adult standard spleen volume (SSV) prediction model based on key physiologic factors. Patients and Methods: A 3D V-Net spleen segmentation model was trained using Dataset 1 (training, n = 3418; validation, n = 413; test, n = 443). Internal validation used Dataset 2 from our institution (n = 1996; January-April 2024), and external validation used 2809 publicly available thin-slice CT examinations. Model performance was evaluated using the Dice similarity coefficient (DSC), Hausdorff distance (HD), and volume similarity (VS). For normative modeling, spleen volumes from 578 healthy adults on portal venous phase CT were analyzed, and candidate prediction variables were assessed using correlation and regression analyses. Results: Segmentation performance was high on the test set {DSC, 0.988 (0.984-0.989) [median (IQR)]; VS, 0.997 (0.994-0.999); HD, 0.017 (0.014-0.021)} and remained robust on external validation [DSC, 0.982 (0.974-0.987)]. Volume agreement analyses showed a mean absolute error of 1.945 mL (test set), 3.829 mL (internal validation), and 5.806 mL (external validation). In the cohort of 578 adults, the spleen volumes ranged from 51.09 to 64438 cm(3), with a median of 177.90 cm(3). The median (IQR) of the x-, y-, and z-axis diameters were 8.71 (7.97-9.53) cm, 9.20 (8.20-10.46( cm, and 9.10 )8.00-1030( cm, respectively. Age (r =-0.24, P < 0.0001) and gender (male = 0, female = 1; r =-032, P < 0.0001) were negatively correlated with splenic volume (SV), while height (r = 0.35, P < 0.0001), weight (W; r = 0.45, P < 0.0001), Body Mass Index (BMI; r = 032, P < 0.0001), and body surface area (BSA; r = 0.46, P < 0.0001) were positively correlated with SV. Based on thin-slice portal venous phase CT images, a bidirectional stepwise selection procedure identified body surface area (BSA) and age as significant predictors. The final model was: log (SSV) = 3.708 + 0.987 x BSA-0.00629 x Age (R-2 = 0.269). Conclusion: A 3D V-Net model enabled accurate automated spleen segmentation with consistent performance across internal and external validation cohorts. An SSV prediction model based on BSA and age provides a population-specific reference for quantitative spleen volumetry in Chinese adults.
Introduction: Aorto-psoas fistulas resulting from infected stent grafts are rare but carry high mortality rates. This case report highlights the critical necessity of distinguishing between a simple abscess and a vascular fistula in patients with prior aortic interventions to prevent catastrophic procedural complications. Case Presentation: An 84-year-old man with a history of lung cancer and prior endovascular aneurysm repair (EVAR) performed 5 years ago presented to the emergency department with severe low back pain. He had a known history of persistent spondylodiscitis and psoas abscess diagnosed 4 months ago, for which he had undergone multiple drainages. The initial admission assessment suggested a recurrent abscess, and percutaneous drainage was arranged. However, the patient's condition rapidly deteriorated with sudden shock and massive gastrointestinal bleeding. An emergency arterial-phase computed tomography angiography (CTA) was performed during resuscitation. The CTA demonstrated contrast medium leakage from the proximal edge of the stent graft, establishing a Type Ia endoleak with fistulous communication to the psoas abscess and the duodenum. Consequently, the contraindicated drainage was withheld. Due to the catastrophic hemorrhage and hemodynamic collapse, the patient died shortly after the diagnosis. Conclusions: This case underscores the necessity of routine contrast-enhanced CTA prior to any image-guided drainage in patients with prior aortic stent grafts. Early identification via CTA is essential to modify management strategies and avoid iatrogenic hemorrhage
Background: Artificial intelligence (Al) systems can improve pulmonary nodule detection, but there is concern that prolonged reliance on AI may alter visual search behavior and affect radiologists' independent interpretive performance when Al support is withdrawn. Objectives: The objective of this study is to evaluate phase-associated changes in pulmonary nodule detection rate after discontinuation of routine Al assistance. Patients and Methods: This retrospective study included 980 chest CT examinations that had been originally interpreted before Al implementation by three senior general radiologists (phase I: Baseline clinical reporting phase, during which they had not previously used any chest CT Al assistance). After approximately 26 months of routine Al use for pulmonary nodule detection, the three participating radiologists discontinued use of the chest CT Al system for this study. Each examination was reassigned to its original reporting radiologist according to report signature and independently reread without Al on the basis of the images alone (phase II), and then reread again in the same manner after a 3-month Al-free washout period (phase III). The pulmonary nodule detection rate, defined as the proportion of scans with at least one reported nodule, and the maximum diameter of the largest reported nodule were compared across phases. Because no external lesion-level reference standard was established, the findings reflect changes in reporting rather than sensitivity or specificity. Results: The pulmonary nodule detection rate decreased from 37.8% (370/980) in phase I to 26.5% (260/980) in phase II and then increased to 43.2% (423/980) in phase III (overall P < 0.0 01). In a generalized estimating equation (GEE) model, using phase I as the reference, the adjusted odds of pulmonary nodule detection were significantly lower in phase II [adjusted odds ratio (aOR) 0.595, 95% confidence interval (CI) 0.530-0.667; P < 0.001] and significantly higher in phase III (aOR1.253, 95% C11.123-1398; P < 0.001). Phase III also showed higher adjusted odds of detection than phase II (aOR 2.106, 95% C11.874-2366; P < 0.001). The phase-related difference was mainly driven by nodules with a maximum reported diameter of <= 5 mm. Conclusion: Discontinuation of routine Al assistance was associated with a short-term decrease in pulmonary nodule detection rate, particularly for small nodules, followed by recovery after an Al-free washout period. These findings suggest a potential vulnerability window during AI downtime or workflow transitions and highlight the need for resilient clinical workflows and performance monitoring.
Background: Non-small cell lung cancer (NSCLC) is a commonly encountered type of lung cancer, accounting for over 85% of lung cancer cases. Objectives: This retrospective cohort study aimed to evaluate the diagnostic differences in time-density curve (TDC) parameters between malignant and benign lung lesions and to explore the associations of these parameters with clinicopathological characteristics and survival outcomes among patients with NSCLC. Methods: Consecutive patients diagnosed between July 2018 and June 2021 were included, comprising 90 pathology-confirmed, treatment-naive NSCLC patients and 90 benign lung lesion controls. All patients underwent multi-slice computed tomography (MSCT) single-slice dynamic scanning, from which contrast enhancement ratio (CER), peak time (PT), and peak value (PV) were extracted. Diagnostic comparisons were performed between groups, and correlations with pathological features were analyzed. Prognostic performance of TDC parameters among NSCLC patients was assessed using receiver operating characteristic (ROC) analysis. Patients were followed for survival outcomes, with a median follow-up duration of 36 months. A Pvalue < 0.05 was considered statistically significant. Results: A total of 90 NSCLC patients (mean age 6135 +/- 6.48 years) were included, with 44 cases at tumor-node-metastasis (TNM) stage I-II and 46 at stage III-IV. Compared with benign lesions, malignant lesions exhibited a significantly delayed time-to-peak, higher peak enhancement, and greater overall enhancement. Within the NSCLC cohort, TDC parameters differed significantly across differentiation grades, TNM stages, and lymph node metastasis status (P < 0.001). The areas under the ROC curves (95% confidence interval) of PT, PV, CER, and their combination for predicting survival outcomes in NSCLC patients were 0.769 (0.657-0.881), 0.776 (0.664-0.888), 0.837 (0.750-0.924), and 0.919 (0.860-0.977), respectively. Conclusions: The MSCT-derived TDC parameters differed significantly between benign and malignant lung lesions and were correlated with pathological characteristics in NSCLC. Importantly, the prognostic associations observed in this study were restricted to the NSCLC cohort, where PT, PV, and CER showed exploratory associations with survival outcome. These findings provide preliminary, hypothesis-generating evidence supporting the potential role of TDC analysis in diagnostic assessment and risk stratification of NSCLC.
Background: Obesity and infertility independently affect testicular function through hormonal disruption, oxidative stress,and endothelial dysfunction. Testicular microvascular perfusion, assessable via Doppler ultrasonography using Resistive Index(RI) and Pulsatility Index (PI), is critical for spermatogenesis and hormonal health. The combined effect of obesity and fertilitystatus on testicular vascular and morphological parameters has not been systematically studied. Objectives: This study aimed to evaluate the independent effects and statistical interaction between obesity and infertility ontesticular microvascular function and volume. Methods: A total of 172 men aged 25 - 45 years were recruited and stratified into four groups: Non-obese fertile (n = 64), non-obese infertile (n = 22), obese fertile (n = 45), and obese infertile (n = 41). Testicular volume and vascular parameters (RI, PI) weremeasured using color Doppler ultrasonography. Non-parametric Kruskal-Wallis tests with post-hoc comparisons, robust linearregression for interaction effects, and Spearman correlations were used for statistical analysis. Results: The Obese-Infertile group had the highest median RI (0.77) and PI (1.61) and the lowest median testicular volume (9.8mL). PI and testicular volume differed significantly across groups (P < 0.001), while RI showed borderline significance (P =0.054). Linear regression revealed a significant statistical interaction between obesity and infertility on PI (beta = 0.318, P < 0.001),indicating that the association with higher PI was stronger in obese infertile men than would be expected from either factoralone. Obese men with preserved fertility maintained relatively normal testicular volume and vascular indices. Spearmancorrelation demonstrated a positive association between BMI and PI and a negative association between BMI and testicularvolume (both P < 0.001). Conclusions: The combination of obesity and infertility was associated with the highest PI values and the lowest testicularvolume among the groups studied. Pulsatility Index and testicular volume are the most sensitive indicators of this dysfunction,while RI is less informative. Obese men who remain fertile maintain relatively preserved testicular volume and vascular indices.
Background: The canalis sinuosus (CS) is an intraosseous canal transmitting the anterior superior alveolar neurovascular bundle. Despite its clinical significance, it is often overlooked. Objectives: This study aimed to evaluate the diameter, location, and distances of the CS relative to adjacent structures using cone-beam computed tomography(CBCT). Methods: This retrospective cross-sectional study analyzed 580 CBCT scans. Multiplanar reconstructions were used to measure canal diameter (CS-D), distance to the anterior nasal spine-posterior nasal spine axis (CS-ANS/PNS), buccal cortical plate (CS-BCP), alveolar crest (CS-AC), and adjacent tooth apices (CS-ATA). CBCT records were evaluated using OnDemand 3D software version 10.0.1. Statistical analyses included chi-square tests, independent t-tests, and generalized estimating equations (GEE). Analyses were adjusted for key confounders. Results: CSs were observed in 219 patients (37.8%), totaling 337 canals. The lateral incisor region was the most common location (44.6%), and the first premolar region the least common (3.6%). Mean CS-D was 0.89 +/- 038 mm, mean CS-ANS/PNS 14.92 +/- 335 mm, mean CS-BCP 7.44 +/- 1.16 mm, mean CS-AC 8.24 +/- 234 mm, and mean CS-ATA 3.83 +/- 138 mm. Canal position significantly influenced all measured distances (P < 0.001). In contrast, laterality (right vs. left) showed a statistically significant difference only for CS-BCP (P = 0.005), with slightly greater values on the right. Conclusion: The CS showed significant anatomical variability. Preoperative CBCT evaluation is essential to minimize injury and ensure predictable surgical outcomes. Radiologists must identify and localize CS and locate it meticulously in presurgical radiological reports of the anterior maxilla.
Background: Artificial intelligence (AI) is rapidly transforming radiology worldwide, yet its adoption and perception amongTurkish radiologists remain underexplored. Objectives: This study aimed to assess the frequency of AI tool usage among Turkish radiologists and to explore theirperspectives on AI's future role in radiology. Patients and Methods: A cross-sectional survey was conducted among 244 practicing radiologists across Turkey. Thequestionnaire collected data on demographics, AI knowledge and training, clinical use of AI tools, perceptions of AI's usefulnessand reliability, and legal and ethical considerations. Descriptive statistics and correlation analyses were performed. Results: Most participants reported basic knowledge of AI (59.8%), with only 38.1% having used AI tools clinically. A strongpositive correlation was observed between AI knowledge and willingness to integrate AI into daily practice (rho = 0.64, P < 0.001).The majority (79.9%) anticipated major changes in radiology due to AI within 10 years and believed AI would reduce workload(76.6%). Formal AI training was weakly correlated with the perceived reliability of AI tools (rho = 0.175, P = 0.006). Legalresponsibility for AI errors was predominantly attributed to software developers (61.1%). Conclusion: Turkish radiologists are optimistic about AI's future in radiology but have limited clinical experience andknowledge. Targeted education and clear regulatory frameworks are essential to support effective and trusted AI integration
Introduction: Inflammatory pseudotumors (IPTs) of the heart are rare, non-neoplastic lesions that can mimic heart neoplasms and are exceptionally uncommon following congenital heart surgery. We report a unique case of a cardiac IPT presenting decades after atrial septal defect (ASD) repair, managed with surgical debulking and targeted therapy. Case Presentation: A 35-year-old woman with a history of ASD repair at age 6 presented in March 2023 with progressive exertional dyspnea over 5 months. Imaging revealed a 4.2 x 3.8 cm mass involving the interatrial septum (IAS), right atrium (RA), and superior vena cava (SVC). Debulking surgery in August 2023 confirmed an IPT histopathologically. Due to residual disease, she received sirolimus and bevacizumab, resulting in a reduction of the residual mass (2.1 x 1.8 cm) and complete resolution of symptoms by August 2024. Conclusions: This case demonstrates the first successful use of sirolimus-bevacizumab combination therapy for a cardiac IPT following ASD repair. This strategy offers a viable alternative for unresectable lesions and underscores the need for long-term surveillance in patients with prior cardiac surgery to detect rare inflammatory complications.
Background: Friedreich's ataxia (FA) is a hereditary neurodegenerative disease frequently complicated by cardiomyopathy, which significantly contributes to patient morbidity and mortality. Conventional imaging techniques may miss early myocardial changes. Objectives: This case series highlights the novel application of advanced cardiac magnetic resonance (CMR) imaging in detecting subclinical myocardial fibrosis and stratifying cardiac risk in FA patients - capabilities often limited in echocardiography. Patients and Methods: Four genetically confirmed FA patients (aged 8 - 22 years) underwent comprehensive CMR, including cine imaging, late gadolinium enhancement (LGE), and tissue mapping sequences. Results: All patients demonstrated concentric left ventricular hypertrophy (LVH), with wall thicknesses ranging from 13 to 21 mm. The LGE revealed patchy myocardial fibrosis in three patients, while Ti mapping indicated diffuse fibrosis even in segments without visible LGE. T2 mapping ruled out active inflammation. Systolic dysfunction [left ventricle ejection fraction (LVEF) 46%] was observed in three cases, while one had preserved function. Conclusion: The CMR provides a unique advantage in identifying and characterizing early cardiac involvement in FA, particularly subclinical fibrosis undetectable by echocardiography. These findings support the integration of CMR into routine FA assessment to improve early diagnosis, risk stratification, and management of cardiomyopathy.
Background: Dental implants can produce artifacts in magnetic resonance imaging (MRI), reducing image quality and diagnostic accuracy. The severity of artifacts can vary depending on implant characteristics and the MRI sequence used. Metal artifact reduction sequences (MARS) have been developed to mitigate artifacts, but their effectiveness across different implant types and sequences requires further investigation. Objectives: To quantify MRI artifacts from titanium (Ti), titanium-zirconium (Ti-Zr), and zirconia (Zr) implants across multiple sequences, with and without MARS, in the anterior and posterior maxilla using a dry human skull model. Materials and Methods: Implants were embedded in agar within a dry human skull at anterior (left lateral maxilla) and posterior (left second molar) positions. The MRI scans comprised ten sequences: T1-weighted (T1W), T2-weighted (T2W), proton density-weighted (PDW), and their 3-dimensional (3D) variants, each performed with and without MARS. Artifact volumes (signal loss and pile-up) were measured using Imalytics Preclinical software. Statistical analysis included ANOVA and Tukey's HSD test (alpha = 0.05). Results: Artifact volumes varied significantly by implant material (P < 0.001), being largest for Ti, intermediate for Ti-Zr, and smallest for Zr. The MARS significantly reduced artifacts in metallic implants (Ti and Ti-Zr, P < 0.001). Importantly, MARS paradoxically increased artifacts in Zr implants. Without MARS, PDW sequences produced the fewest artifacts, while 3D T1W sequences generated the most (P = 0.03). Anterior implants showed greater signal loss than posterior implants (P < 0.001), with slightly higher pile-up, whereas total artifact differences were not significant (P = 0.263). Conclusion: Artifact severity strongly depends on implant material, with Ti producing the most and Zr the least artifacts. The PDW imaging minimizes artifact extent. While MARS effectively reduces artifacts in metallic implants, it paradoxically worsens artifacts in Zr, emphasizing the need for careful protocol selection for non-metallic implants.
Background: Flow diverter (FD) devices have emerged as a promising option for treating intracranial aneurysms, particularly those that are complex or wide-necked. Although their efficacy in achieving aneurysm occlusion is well established, real-world data on complication rates remain limited. Objectives: The objective of this study was to describe the incidence and characteristics of complications following FD treatment of intracranial aneurysms in a single-center, real-world cohort. Patients and Methods: This retrospective, single-center study included 27 patients who underwent a total of 29 FD procedures for intracranial aneurysms. Demographic data, aneurysm morphology, and procedural details were collected. Due to the limited sample size and descriptive focus, no comparative statistical analyses were conducted. Continuous variables were summarized using means and standard deviations; categorical variables were presented as frequencies and percentages. Complication and mortality rates were reported with 95% confidence intervals (CIs) to reflect statistical uncertainty. Results: The mean patient age was 49.4 +/- 12.8 years (range: 18-82), and 70.4% were female. Aneurysm types included saccular (793%), dissecting (13.8%), fusiform with wide neck (3.4%), and pseudoaneurysm (3.4%). Adjunctive coiling was used in 55.2% of procedures, with an average of 2.6 +/- 2 coils per procedure. The overall complication rate was 13.8% (4/29; 95% CI: 3.9-31.7%) including stent migration, in-stent thrombosis, intracerebral hemorrhage, and complete carotid artery thrombosis. The mortality rate was 10.3% (3/29; 95% CI: 2.2-27.4%). One patient experienced permanent visual loss. Complete aneurysm occlusion was achieved in 82.8% of cases during follow-up. Due to the small number of patients treated with non-flow re-direction endoluminal device (FRED) devices, no subgroup comparisons were performed. Conclusion: These findings highlight the importance of close procedural monitoring and underscore the need for larger, prospective studies to further assess FD-related complications and long-term outcomes. Further studies with larger populations and longer follow-up periods are necessary to compare the efficacy and safety of different FDs.
Background: Chest X-ray (CXR) images are important for diagnosing lung diseases such as pneumonia and tuberculosis. They help medical professionals determine the functions of the heart and lungs. The lungs may change as a result of certain heart issues, and specific disorders may cause anatomical alterations of the heart or lungs. Objectives: To enhance the classification accuracy of CXR images, particularly for identifying eight types of abnormal lung lesions, by applying advanced feature selection and fusion techniques in combination with deep learning models. Materials and Methods: This study utilized a dataset of CXR images with normal and abnormal classes; however, we had only eight lesions in the abnormal class. Initially, we preprocessed the images to improve their quality and suitability for further analysis. We then modified two deep learning models, ResNet50 and DenseNet201. Transfer learning (TL) techniques were employed for training. The features extracted from these models were retained separately. We utilized a parameter-optimized ant colony optimization (ACO) algorithm to refine the feature selection, and the features from both models were fused into a single feature set. Meanwhile, at the preprocessing step, we used five additional statistical methods, including the Kruskal-Wallis test, ReliefF, analysis of variance (ANOVA), chi-square test, and minimum redundancy maximum relevance (MRMR), to identify the most important features, separately from the above process. The fusion is then enforced after obtaining important features from two different processes to enhance the efficiency of our architecture. We then utilized various machine learning classifiers. Results: The binary classification between normal and abnormal achieved 95.9% +/- 0.5% accuracy, 95.95% +/- 1.09% sensitivity, 93.91% +/- 034% specificity, 93.91% +/- 037% precision, and 94.85% +/- 0.56% Fl score on the hybrid approach. Multiple classifications of the eight abnormality lesions revealed a promising average area under the curve (AUC) value of 0.872. Conclusion: The combination of deep learning models with advanced feature selection methods is beneficial for not only improving the classification outcomes but also ensuring accuracy and validity.
Background: During pregnancy, numerous changes occur in thyroid gland functions. Ultrasonography (US) is one of the primary imaging methods used for thyroid gland examination. Two-dimensional shear wave elastography (2D-SWE) is an imaging method that allows the quantitative evaluation of the stiffness of tissues, and its use in thyroid gland diseases is gradually increasing. Objectives: The present study aimed to investigate whether there are changes in thyroid parenchymal stiffness during pregnancy and to determine normal 2D-SWE values of the thyroid gland in each trimester. Methods: Forty-eight healthy pregnant women were evaluated. The dimensions of thyroid lobes were measured separately, and volume values were calculated. The 2D-SWE measurements were made by drawing a circular region of interest (ROI) in the axial plane for both thyroid lobes. Results: The mean age was 28.4 +/- 6.0 years. The mean elasticity values (kPa) for the three trimesters were 8.77 +/- 1, 8.1 +/- 0.7, and 7.6 +/- 0.7. Mean velocity values (m/s) were 1.71 +/- 0.1, 1.64 +/- 0.08, and 1.60 +/- 0.07. Total thyroid volume values (mL) were 9.72 +/- 4.21, 10.69 +/- 4.55, and 11.75 +/- 5.2. The 2D-SWE elasticity and velocity values significantly decreased across trimesters (P < 0.05), whereas thyroid volume showed a non-significant trend toward increase. Conclusion: Our study is one of the few longitudinal prospective studies in the literature in which the thyroid gland was evaluated sonographically in pregnant women. We quantitatively demonstrated, with objective numerical data, a significant decrease in thyroid elastography values during pregnancy, while thyroid volume exhibited a non-significant tendency to increase.
