Current Medical Imaging - Current Issue

The study aimed to evaluate the clinical efficacy of CT-guided radiofrequency ablation in conjunction with targeted therapy in lung cancer patients.

We retrospectively analyzed 80 lung cancer patients. They were stratified into the Observation Group (OG; n=40, treated with CT-guided radiofrequency ablation in conjunction with targeted therapy) and the Control Group (CG; n=40, treated solely with targeted therapy).

The Overall Response Rate (ORR) and Disease Control Rate (DCR) in the OG group (70.00%, 95.00%) were significantly higher than those in the CG group (57.50%, 87.50%). After 6 weeks of treatment, the OG group had significantly lower levels of SCC, CEA, and CA125, higher CD4+ levels, and lower CD8+ levels, compared to the CG group. The 24-month follow-up survival rate of the OG group (47.50%) was significantly higher than that of the CG group (27.50%).

CT-guided radiofrequency ablation and targeted therapy have been proven effective in treating lung cancer.

There remains a lack of methods to accurately assess the efficacy of neoadjuvant chemoradiotherapy for locally advanced rectal cancer.

This study aimed to investigate the value of shear-wave elastography in evaluating the treatment response to neoadjuvant chemoradiotherapy for locally advanced rectal cancer.

This prospective observational study enrolled 275 patients with locally advanced rectal cancer who received neoadjuvant chemoradiotherapy during September 2021–March 2023. All patients underwent endorectal ultrasound and shear-wave elastography examination before total mesorectal excision. Clinical baseline data, endorectal ultrasound, and shear-wave elastography examination data were collected from all patients. The independent predictors of complete response were analyzed and screened, followed by the establishment of a logistic regression model. The diagnostic efficacy of the model was compared with that of radiologists.

The results of binary multivariate logistic regression suggested that the mean elastography value of the tumor lesion acted as an independent predictor for the diagnosis of complete response [OR: 0.894 (0.816, 0.981)]. The optimal cutoff value was 14.6 kPa. The area under the receiver operating characteristic curve of the model for predicting complete response in the training and test cohorts was 0.850 and 0.824, respectively. The diagnostic accuracy of the model was significantly higher than that of radiologists (P < 0.001).

Shear-wave elastography can be used as a feasible method to evaluate the complete response of locally advanced rectal cancer after neoadjuvant chemoradiotherapy.

Glaucoma is a leading cause of irreversible visual impairment and blindness worldwide, primarily linked to increased intraocular pressure (IOP). Early detection is essential to prevent further visual impairment, yet the manual diagnosis of retinal fundus images (RFIs) is both time-consuming and inefficient. Although automated methods for glaucoma detection (GD) exist, they often rely on individual models with manually optimized hyperparameters. This study aims to address these limitations by proposing an ensemble-based approach that integrates multiple deep learning (DL) models with automated hyperparameter optimization, with the goal of improving diagnostic accuracy and enhancing model generalization for practical clinical applications.

The RFIs used in this study were sourced from two publicly available datasets (ACRIMA and ORIGA), consisting of a total of 1,355 images for GD. Our method combines a custom Multi-branch convolutional neural network (CNN), pretrained MobileNet, and DenseNet201 to extract complementary features from RFIs. Moreover, to optimize model performance, we apply Bayesian Optimization (BO) for automated hyperparameter tuning, eliminating the need for manual adjustments. The predictions from these models are then combined using a Dirichlet-based Weighted Average Ensemble (Dirichlet-WAE), which adaptively adjusts the weight of each model based on its performance.

The proposed ensemble model demonstrated state-of-the-art performance, achieving an accuracy (ACC) of 95.09%, precision (PREC) of 95.51%, sensitivity (SEN) of 94.55%, an F1-score (F1) of 94.94%, and an area under the curve (AUC) of 0.9854. The innovative Dirichlet-based WAE substantially reduced the false positive rate, enhancing diagnostic reliability for GD. When compared to individual models, the ensemble method consistently outperformed across all evaluation metrics, underscoring its robustness and potential scalability for clinical applications.

The integration of ensemble learning (EL) and advanced optimization techniques significantly improved the ACC of GD in RFIs. The enhanced WAE method proved to be a critical factor in achieving well-balanced and highly accurate diagnostic performance, underscoring the importance of EL in medical diagnosis.

Lipomas represent a rare benign etiology of intussusception in adults, affecting both the small intestine and the colon. Diagnosing intussusception in adults can be challenging, and there are no reports on the use of positron emission tomography/CT (PET/CT) in the diagnosis of lipoma-induced intussusception. This study aimed to preliminarily explore the potential diagnostic utility of ¹⁸F-FDG PET/CT in the diagnosis of intussusception caused by lipomas.

We conducted a retrospective review of the clinical characteristics and imaging findings of three patients diagnosed with lipoma-induced intussusception using ¹⁸F-FDG PET/CT from 2019 to 2023 at our hospital.

The three cases presented with diverse clinical presentations and were diagnosed based on PET/CT imaging. Surgical confirmation was obtained in two cases. Lipomas were identified in both the small intestine and the colon, with one case displaying increased metabolic activity on FDG uptake, suggesting a possible link between FDG uptake and clinical severity.

Lipoma is a benign cause of intussusception that can occur in both the small intestine and the colon. The symptoms of adult intussusception are often atypical and variable. Imaging modalities, particularly PET/CT, are instrumental in diagnosing intussusception due to lipomas, differentiating between benign and malignant causes, and assessing the severity to inform treatment strategies.

MRI could be considered as a non-destructive disease diagnosis procedure, this procedure does not allow directly molecular types of cancer. Herein, we aimed to evaluate the correlation of breast MRI background parenchymal enhancement (BPE) and fibroglandular tissue (FGT) with the molecular subtypes and immunohistochemical markers of breast cancer.

This was a single-cross-sectional retrospective study.Fifty-six patients diagnosed with unilateral breast cancer who underwent breast MRI scans before needle biopsy or surgery were selected. The relationship between qualitative and quantitative BPE/FGT ratios and the expression of breast cancer molecular subtypes and immunohistochemical markers were evaluated in patients with breast cancer.

Quantitative BPE (BPE%) of luminal A and luminal B was significantly lower than that of triple-negative breast cancer. There was no significant difference in the qualitative BPE/FGT between the different breast cancer subtypes. The quantitative BPE (BPE%) of estrogen receptor (ER)-negative tumors was higher than that of the ER-positive tumors, and the expression of FGT%, BPE%, and other immunohistochemical markers (human epidermal growth factor receptor-2(HER-2), progesterone receptor (PR), and Ki-67) were not significantly different. The proportion of high BPE distribution in HER-2 positive tumors was higher than that in the HER-2 negative group; however, there was no significant difference in the expression of qualitative BPE/FGT and other immunohistochemical markers (ER, PR, and Ki-67).

There were significant differences in the levels of BPE among the different molecular subtypes. Therefore, BPE may be a potential imaging biomarker for the diagnosis of the molecular subtypes of breast cancer.

Although the “conventional” autopsy is still considered the “gold standard” for the definition of the cause of death, an emerging interest in non-invasive cadaveric investigations is spreading. Among all these, the application of post-mortem magnetic resonance imaging of the heart is increasingly gaining ground in the study of sudden cardiac death.

Using the diffusion tensor imaging sequence, the present study aimed to demonstrate how through the fractional anisotropy value it is possible to qualitatively and quantitatively define sudden cardiac death, particularly in cases of sudden arrhythmic death syndrome. Four hearts were collected for the present pilot study: the first from a subject who died from a brain injury caused by a gunshot, and the other three hearts from subjects who died of sudden cardiac death. In all cases examined, the extracted hearts were hung inside a container containing 10% formalin solution and placed inside a 1.5T scanner with a 16-channel chest coil. Then, the cardiac diffusion tensor imaging sequence was performed and the quantitative maps of fractional anisotropy and apparent diffusion coefficient were obtained. After imaging analysis, the samples were processed, paraffin-embedded, and stained with hematoxylin and eosin and trichrome staining. Cases B, C, and D showed lower fractional anisotropy values than non-pathological one.

Histological investigation revealed extensive areas of fibrosis and foci of contraction band necrosis in heart B, myofiber disarray and interstitial fibrosis in heart C, and findings consistent with atonic death in heart D.

The study aimed to demonstrate that in cases of sudden cardiac death, lower fractional anisotropy values, as already observed in clinical trials, are associated with arrhythmic heart disease or myocardial infarction. Quantitative, appreciable, and reproducible data could support such diagnoses.

In this case series, we present two male cases with renal parenchymal perforation without perirenal hematoma after D-J ureteral stent insertion, one with nutcracker renal vein syndrome. Our study provides new and important contributions to the field of science regarding what to consider during D-J stent insertion in similar cases and in patients with obstruction in the urinary collecting system for more than 2 months.

Two patients, 30 and 37 years old, who were inserted a D-J catheter after endoscopic ureteral stone treatment, suffered from severe ipsilateral flank pain and hematuria after the operation. The Kidney Urine Bladder (KUB) radiography showed that the proximal part of the D-J stent was protruding from the upper calyx and parenchyma of the kidney in both patients. One of the patients had an ipsilateral nutcracker renal vein syndrome, and the clinical progression was more severe. In both cases, conventional follow-up with bed rest, nonsteroidal anti-inflammatory (NSAI) therapy, intravenous (IV) fluid infusion, and anti-biotherapy after the D-J stent reposition was sufficient. The patients had no clinical problems during the next outpatient clinic visits.

Double-j (D-J) ureteral stent insertion procedure may cause many life-threatening complications, from subcapsular hematoma to pulmonary embolism. In this case series, conventional follow-up was sufficient for the treatment of patients with renal parenchymal damage without perirenal hematoma due to D-J stent insertion, including nutcracker renal vein syndrome cases. More care should be taken when placing D-J stents, especially in patients with obstruction in the urinary collecting system for more than 2 months and with nutcracker renal vein syndrome. In these patients, the soft proximal end of the guidewire should not be pushed and forced too hard to the upper part of the kidney and upper collecting system. Additionally, the D-J stent placement procedure should be performed under fluoroscopy as much as possible to avoid complications.

The purpose of this study was to investigate the morphology and distribution characteristics of fat globules in osteomyelitis on magnetic resonance imaging (MRI).

Patients with pathologically-confirmed osteomyelitis and MRI scans were retrospectively enrolled, and fat globules on the MRI images were analyzed.

Among 103 patients with non-traumatic osteomyelitis, 75 were fat globule negative and 28 were positive. There was no statistically significant difference in age and gender between patients with and without fat globules (p>0.05). The inflammatory indicators (CRP, ESR, WBC, and NEUT) in the fat globule positive group were significantly higher (p<0.05) than those in the negative group. The lesions were mainly located in the long bones of the limbs in patients with positive fat globules. Twenty-eight patients (27.2% or 28/103) were detected to have fat globules on MRI images, including 20 males (71%) and 8 females (29%) aged 5-64 years (mean 16 years). The time from onset to MRI examination was 8 days to 4 months. The location of fat globules was in the tibia in 10 patients (35.7%), femur in 8 (28.6%), humerus in 4 (14.3%), radius in 2 (7.1%), ulna in 1 (3.6%), calcaneus in 1 (3.6%), sacrum in 1 (3.6%), and fibula in 1 patient (3.6%). On MRI imaging, 28 cases (100%) showed widely distributed patches or tortuous and sinuous abnormal signals in the bone marrow. In 25 cases (89.2%), a grid-like abnormal signal was found in the subcutaneous soft tissue. In 21 patients (75%), pus was found in the adjacent extraosseous soft tissues. Among 28 patients with fat globules, 17 patients (60.7%) had fat globules only in the adjacent extraosseous soft tissue, 6 patients (21.4%) had only intraosseous fat globules (including 5 cases with halo signs around the fat globules and 1 case (3.6%) with fat globules located at the edge of the pus cavity inside the bone without a halo sign), and 5 patients (17.8%) had both intraosseous and extraosseous fat globules. Of 6 patients (21.4% or 6/28) with liquid levels, the liquid level appeared outside the bone.

The appearance of fat globules on MRI in patients with osteomyelitis indicates severe infection. Fat globules of osteomyelitis may present with diverse shapes inside and outside the bone marrow as one of the MRI signs of osteomyelitis, with a probability of approximately 27.2%. They have high specificity in diagnosing osteomyelitis and can be used for diagnosis and differential diagnosis.

Delphian lymph node (DLN) metastasis is common in papillary thyroid cancer (PTC). However, few studies have specifically investigated the clinicopathologic characteristics of DLN metastasis in PTC. This study aimed to examine the incidence, risk factors, and predictive value of DLN in papillary thyroid carcinoma.

In the present study, the clinicopathologic features and metastatic risks were statistically analyzed by reviewing 1837 patients with papillary thyroid carcinoma who underwent initial surgery in our department between January, 2022 and July, 2024.

Among the 1837 patients included in the study, DLN was detected in 925 patients (50.3%), of which 409 patients (22.3%) had confirmed DLN metastasis. In univariate analysis, DLN metastasis was correlated with age (≥55 years), bilateral cancer, multifocality, tumor location (isthmus cancer), central lymph node metastasis (CLNM), and lateral lymph node metastasis (LLNM). However, it was not correlated with gender distribution, tumor size, thyroiditis, thyroid-stimulating hormone (TSH) level, and BRAF mutation. Multivariate analysis showed that CLNM (p=0.03), LLNM (p=0.025), bilateral cancer, and tumor location (p=0.012) were independent risk factors for DLN metastasis. DLN involvement was mildly predictive of CLNM (sensitivity=29.76%, specificity=87.06%, positive predictive values=74.08%, negative predictive values=49. 93%, positive likelihood ratio=2.30, negative likelihood ratio=0.81) and moderately predictive of LLNM (sensitivity=49.36%, specificity=85.01%, positive predictive values=46.94%, negative predictive values=86.20%, positive likelihood ratio=3.29, negative likelihood ratio=0.60).

Bilateral cancer, CLNM, LLNM, and isthmus cancer were independent risk factors for DLN metastasis. DLN metastasis could be used as a predictor for central and lateral lymph node metastasis. Positive DLN should be a warning signal to carefully evaluate central and lateral lymph nodes during thyroidectomy.

This study aimed to determine the relationship between Radiation-Induced Lung Injury (RILI) and the clinical outcome of Non-Small Cell Lung Cancer (NSCLC) following Stereotactic Ablative Radiotherapy (SABR).

Clinical data and follow-up CT scanning of 101 patients with early NSCLC who received SABR treatment from January 2012 to December 2018 were retrospectively collected, and the Progression Free Survival (PFS) was calculated. CT features of peritumoral RILI were observed by 3 radiologists, each with 10 to 15 years of experience, based on consensus among 3 radiologists and divided into 3 types. Type 1: Diffuse consolidation surrounding the tumor, including the tumor boundary. Type 2: Ground Glass Opacities (GGOs) covering more than 180 degrees around the tumor. Type 3: GGOs surrounding the tumor but covering less than 180 degrees. Log-rank test was used to analyze the correlation between the classification of radiation-induced lung injury CT findings and PFS. Independent predictors of PFS rate were analyzed by COX multivariate regression.

The 5-year PFS rates based on RILI types observed at 6-8 months post-SABR were: Type 1 = 69.5%, Type 2 = 50.9%, and Type 3 = 36.1%. A statistically significant difference was observed among the three RILI types (p=0.025). COX multivariate regression analysis showed that RILI were independent factors influencing PFS (at 6-8 months follow-up after radiotherapy (p=0.041).

Patients with more extensive and denser RILI tend to have a longer PFS. Data from our cohort study indicate that the 6- 8 months post-SABR period represents the optimal follow-up window, as evidenced by significant progression-free survival rate dynamics during this interval (HR = 1.5, 95% CI 1.0-2.2, p < 0.05).

This study seeks to assess vasculogenic mimicry (VM) occurrence in lung adenocarcinoma (LUAD) by delineating intratumoral and peritumoral characteristics using preoperative CT-based radiomics and a nomogram for enhanced precision.

Our retrospective analysis enrolled 150 LUAD patients, ascertained their VM status, and stratified them randomly into development (n=105) and validation cohorts. We extracted radiomics features from intra- and peritumoral zones, delineating 3, 5, and 7mm expansions on thin-section chest CT images. We formulated logistic models encompassing a clinical model (CM), intratumoral radiomics model (TRM), peritumoral radiomics models at 3, 5, and 7 mm (PRMs), and a composite model integrating both intra- and peritumoral zones (CRM). A radiomics nomogram model (RNM) was devised, amalgamating the Rad-scores from intra- and peritumoral regions with clinical-radiological traits to forecast VM. The models' efficacy was gauged via the receiver operating characteristic (ROC) curve analysis, calibration assessment, and decision curve analysis (DCA).

The CRM outperformed its counterparts, the TRM, PRM_3mm, PRM_5mm, and PRM_7mm models, with AUCs reaching 0.859 and 0.860 in the development and validation cohorts. Within the CM, tumor size and spiculation emerged as significant predictive covariates. The RNM, integrating independent predictors with the CRM-Rad-score, demonstrated clinical utility, achieving AUCs of 0.903 and 0.931 in the respective cohorts.

Our findings underscore the potential of CT-based radiomics characteristics derived from intratumoral and peritumoral regions to assess VM presence in LUAD patients. Combining radiomics signatures with clinicoradiological parameters within a nomogram framework significantly enhances predictive accuracy.

This study aimed to assess the hemodynamic changes in the vena cava and predict the likelihood of Cardiac Remodeling (CR) and Myocardial Fibrosis (MF) in athletes utilizing four-dimensional (4D) parameters.

A total of 108 athletes and 29 healthy sedentary controls were prospectively recruited and underwent Cardiac Magnetic Resonance (CMR) scanning. The 4D flow parameters, including both general and advanced parameters of four planes for the Superior Vena Cava (SVC) and Inferior Vena Cava (IVC) (sheets 1-4), were measured and compared between the different groups. Four machine learning models were employed to predict the occurrence of CR and/or MF.

Most general 4D flow parameters related to VC were increased in athletes and positive athletes compared to controls (p < 0.05). Gradient Boosting Machine (GBM) was the most effective model in sheet 2 of SVC, with the area under the curve values of 0.891, accuracy of 85.2%, sensitivity of 84.6%, and specificity of 85.4%. The top five predictors in descending order were as follows: net positive volume, forward volume, waist circumference, body weight, and body surface area.

Physical activity can induce a high flow state in the vena cava. CR and/or MF may elevate the peak velocity and maximum pressure gradient of the IVC. This study successfully constructed a GBM model with high efficacy for predicting CR and/or MF. This model may provide guidance on the frequency of follow-up and the development of appropriate exercise plans for athletes.

Fibrosing mediastinitis (FM) is a rare and benign disease affecting the mediastinum and often causes pulmonary hypertension (PH). Timely diagnosis of PH caused by FM is clinically important to mitigate complications such as right heart failure in affected individuals. This retrospective study aimed to analyze the CT imaging characteristics of tuberculosis (TB) related FM in patients with (TB). Additionally, the study investigates the underlying reasons contributing to the manifestation of symptoms.

From April 2007 to October 2020, high-resolution CT (HRCT) and dual-phase CT pulmonary angiography images of 64 patients with suspected FM diagnosed with PH at a tertiary hospital were examined. The imaging characteristics of these CT scans were analyzed, with a specific focus on the TB-FM involvement of the pulmonary veins, pulmonary arteries, and bronchi (down to the segment level).

HRCT imaging revealed that fibrous tissue inside the mediastinum exhibited minimal or negligible reinforcement in TB-FM and diffuse fibrous infiltration in the mediastinum and hilar areas. Notably, segmental bronchial and pulmonary artery stenosis are more pronounced and frequently co-occurring than lobe-level stenosis. Pulmonary venous stenosis developed outside the pericardium, whereas pulmonary artery stenosis occurred outside the mediastinal pleura. Furthermore, no isolated FM involvement in pulmonary veins was noticed in this cohort.

HRCT imaging of TB-related FM presents unique features in certain regions of the bronchi, pulmonary veins, and pulmonary arteries. Thus, it is imperative to accurately identify fibrous tissue involvement in mediastinal lesions for proper diagnosis and management of TB-FM.

The aim of this study was to develop and validate predictive models for perineural invasion (PNI) in gastric cancer (GC) using clinical factors and radiomics features derived from contrast-enhanced computed tomography (CE-CT) scans and to compare the performance of these models.

This study included 205 GC patients, who were randomly divided into a training set (n=143) and a validation set (n=62) in a 7:3 ratio. Optimal radiomics features were selected using the least absolute shrinkage and selection operator (LASSO) algorithm. A radiomics model was constructed utilizing the optimal among five machine learning filters, and a radiomics score (rad-score) was computed for each participant. A clinical model was built based on clinical factors identified through multivariate logistic regression. Independent clinical factors were combined with the rad-score to create a combined radiomics nomogram. The discrimination ability of the models was evaluated by receiver operating characteristic (ROC) curves and the DeLong test.

Independent predictive factors of the clinical model included tumor T stage, N stage, and tumor differentiation, with AUC values of 0.777 and 0.809 in the training and validation sets. The radiomics model was constructed using the support vector machine (SVM) classifier with the best AUC (0.875 in the training set and 0.826 in the validation set). The combined radiomics nomogram, which combines independent clinical predictors and the rad-score, demonstrated better predictive performance (AUC=0.889 in the training set; AUC=0.885 in the validation set).

The nomogram integrating independent clinical predictors and CE-CT radiomics was constructed to predict PNI in GC. This model demonstrated favorable performance and could potentially assist in prognosis evaluation and clinical decision-making for GC patients.

Muscle infiltration of bladder cancer has become the most important index to evaluate its prognosis. Machine learning is expected to accurately identify its muscle infiltration status on images.

This study aimed to establish and validate a machine learning prediction model based on multi-phase contrast-enhanced CT (MCECT) for preoperatively evaluating the muscle-invasive status of bladder cancer.

A retrospective study was conducted on bladder cancer patients who underwent surgical resection and pathological confirmation by MCECT scans. They were randomly divided into training and testing groups at a ratio of 8:2; we used an independent external testing set for verification. The radiomics features of lesions were extracted from MCECT images and radiomics signatures were established by dual sample T-test and least absolute shrinkage selection operator (LASSO) regression. Afterward, four machine learning classifier models were established. The receiver operating characteristic (ROC) curve, calibration, and decision curve analysis were employed to evaluate the efficiency of the model and analyze diagnostic performance using accuracy, precision, sensitivity, specificity, and F1-score.

The best predictive model was found to have logic regression as the classifier. The AUC value was 0.89 (5-fold cross-validation range 0.83-0.96) in the training group, 0.80 in the test group, and 0.87 in the external testing group. In the testing and external testing group, the diagnostic accuracy, precision, sensitivity, specificity, and F1-score were 0.759, 0.826, 0.863, 0.729, 0.785, and 0.794, 0.755, 0.953, 0.720, and 0.809, respectively.

The machine learning model showed good accuracy in predicting the muscle infiltration status of bladder cancer before surgery.

Pneumoconiosis is a preventable occupational lung disease that is caused by the inhalation of inorganic occupational dust. The disease can progress and result in functional impairment. Profusion scores are crucial for the assessment of disease severity.

This study aimed to determine the prevalence of pneumoconiosis cases with a profusion score of 0/1 and explore the correlation between pneumoconiosis and smoking behavior and sectors.

A retrospective cross-sectional study was carried out in this work. Pneumoconiosis was diagnosed with occupational exposure histories and thoracic computed tomography (CT) findings. The study included patients admitted to the occupational diseases outpatient clinic at Eskişehir City Hospital for occupational or pulmonary conditions from January 2021 to July 2023. The collected data included age, sex, smoking status, pack-years, industry of employment, specific departments, occupations, exposure to occupational and non-occupational environmental factors, duration of exposure, laboratory results, pulmonary function test outcomes, thoracic CT findings, and International Classification of Radiographs of Pneumoconiosis score.

Among the 361 patients, 99.4% were male and 62.3% were current smokers. We observed a profusion score of 0/1 in 15% (n = 54) of the cases. Patients with a 0/1 profusion score had better lung function than those with higher scores, with the FEV1/FVC ratio declining as the profusion score increased. Non-smokers with progressive massive fibrosis had significantly lower FEV1/FVC ratios compared to other non-smokers.

In order to avert the progression of early-stage cases, it is significant that we reevaluate occupational health policies and measures, regardless of compensation.

The malignant risk of cystic-solid thyroid nodules may be underestimated in the ultrasound assessment.

This systematic review and meta-analysis aimed to evaluate the value of multimodal ultrasound characteristics in the malignant risk assessment of cystic-solid thyroid nodules.

We conducted a comprehensive search of PubMed, Web of Science, and Cochrane Library databases for studies depicting the ultrasound characteristics of cystic-solid thyroid nodules published prior to October 2023. The Review Manager 5.4 software was utilized to evaluate the ultrasound features suggestive of malignancy and to determine their sensitivity and specificity. Additionally, the Sata 12.0 software was utilized to construct summary receiver operating characteristic curves (SROC), estimate the area under the curve (AUC), and evaluate any potential publication bias.

This review included 16 studies comprising 5,655 cystic-solid thyroid nodules. Nine ultrasound features were identified as risk factors for tumor malignancy. Among the ultrasound features, microcalcification in the solid portion, heterogeneous hypoenhancement on Contrast-Enhanced Ultrasound (CEUS), and sharp angles in the solid portion exhibited higher malignant predictive value in cystic-solid thyroid nodules, with AUC values of 0.91, 0.84, and 0.81, respectively.

Our findings indicate that microcalcification and sharp angles in the solid part of the nodule, along with heterogeneous hypoenhancement on contrast-enhanced ultrasound (CEUS), can better predict malignant cystic-solid thyroid nodules.

The systematic review and meta-analysis was registered prospectively in the International Prospective Register of Systematic Reviews (No. CRD42024602893).

The objective of this research is to enhance pneumonia detection in chest X-rays by leveraging a novel hybrid deep learning model that combines Convolutional Neural Networks (CNNs) with modified Swin Transformer blocks. This study aims to significantly improve diagnostic accuracy, reduce misclassifications, and provide a robust, deployable solution for underdeveloped regions where access to conventional diagnostics and treatment is limited.

The study developed a hybrid model architecture integrating CNNs with modified Swin Transformer blocks to work seamlessly within the same model. The CNN layers perform initial feature extraction, capturing local patterns within the images. At the same time, the modified Swin Transformer blocks handle long-range dependencies and global context through window-based self-attention mechanisms. Preprocessing steps included resizing images to 224x224 pixels and applying Contrast Limited Adaptive Histogram Equalization (CLAHE) to enhance image features. Data augmentation techniques, such as horizontal flipping, rotation, and zooming, were utilized to prevent overfitting and ensure model robustness. Hyperparameter optimization was conducted using Optuna, employing Bayesian optimization (Tree-structured Parzen Estimator) to fine-tune key parameters of both the CNN and Swin Transformer components, ensuring optimal model performance.

The proposed hybrid model was trained and validated on a dataset provided by the Guangzhou Women and Children’s Medical Center. The model achieved an overall accuracy of 98.72% and a loss of 0.064 on an unseen dataset, significantly outperforming a baseline CNN model. Detailed performance metrics indicated a precision of 0.9738 for the normal class and 1.0000 for the pneumonia class, with an overall F1-score of 0.9872. The hybrid model consistently outperformed the CNN model across all performance metrics, demonstrating higher accuracy, precision, recall, and F1-score. Confusion matrices revealed high sensitivity and specificity with minimal misclassifications.

The proposed hybrid CNN-ViT model, which integrates modified Swin Transformer blocks within the CNN architecture, provides a significant advancement in pneumonia detection by effectively capturing both local and global features within chest X-ray images. The modifications to the Swin Transformer blocks enable them to work seamlessly with the CNN layers, enhancing the model’s ability to understand complex visual patterns and dependencies. This results in superior classification performance. The lightweight design of the model eliminates the need for extensive hardware, facilitating easy deployment in resource-constrained settings. This innovative approach not only improves pneumonia diagnosis but also has the potential to enhance patient outcomes and support healthcare providers in underdeveloped regions. Future research will focus on further refining the model architecture, incorporating more advanced image processing techniques, and exploring explainable AI methods to provide deeper insights into the model's decision-making process.

In the dynamic landscape of healthcare, integrating Artificial Intelligence paradigms has become essential for sophisticated brain image analysis, especially in tumor detection. This research addresses the need for heightened learning precision in handling sensitive medical images by introducing the Fragmented Segment Detection Technique.

The ever-evolving healthcare landscape demands advanced methods for brain image analysis, particularly in detecting tumors. This study responds to this need by introducing the Feature Segmentation and Detection Technique (FSDT), a novel approach designed to identify brain tumors precisely using MRI images. The focus is on enhancing detection accuracy, even for diminutive tumors.

The primary objective of this study is to introduce and evaluate the efficacy of FSDT in identifying and sizing brain tumors through advanced medical image analysis. The proposed technique utilizes cross-section segmentation and pixel distribution analysis to improve detection accuracy, particularly in size-based tumor detection scenarios.

The proposed technique commences by fragmenting the input through cross-section segmentation, enabling meticulous separation of pixel distribution in various sections. A Convolutional Neural Network then independently operates sequentially on the minimum and maximum representations. The segmented cross-section feature, exhibiting maximum accuracy, is employed in the neural network training process. Fine-tuning of the neural network optimizes feature distribution and pixel arrangements, specifically in consecutive size-based tumor detection scenarios.

The FSDT employs cross-sectional segmentation and pixel distribution analysis to enhance detection accuracy by leveraging a diverse dataset encompassing central nervous system CNS tumors. Comparative evaluations against existing methods, including ERV-Net, MRCNN, and ENet-B0, reveal FSDT's superiority in accuracy, training rate, analysis ratio, precision, recall, F1-score, and computational efficiency. The proposed technique demonstrates a remarkable 10.45% increase in accuracy, 14.12% in training rate, and a 10.78% reduction in analysis time.

The proposed FSDT emerges as a promising solution for advancing the accurate identification and sizing of brain tumors through cutting-edge medical image analysis. The demonstrated improvements in accuracy, training rate, and analysis time showcase its potential for effective real-world healthcare applications.

The earlier detection of cervical cancer in women patients can save human life. This article proposes a novel methodology for detecting abnormal cervigram images from healthy cervigram images and segments the cancer regions in the abnormal cervigram images using the deep learning method. The conventional deep learning architecture has been modified into the proposed CervixNet architecture to improve the cervical cancer detection rate.

This methodology is constituted of a training and testing process, where the training process generates the training sequences individually for healthy cervigram images and the cancer case cervigram images. The testing process tests the cervigram images into either a healthy or cancer cases using the training sequences generated through the training process. During the testing process of the proposed system, the cancer segmentation algorithm was applied on the abnormal cervigram image to detect and segment the pixels belonging to cancer. Finally, the performance has been carried out on the segmented cancer cervical images for the ground truth images. This proposed methodology has been evaluated on the cervigrams on IMODT and Guanacaste databases. Its performance has been analyzed concerning cancer pixel sensitivity, cancer pixel specificity and cancer pixel accuracy.

This research work obtains 98.69% Cancer Pixel Sensitivity (CPS), 98.76% Cancer Pixel Specificity (CPSP), and 99.27% Cancer Pixel Accuracy (CPA) for the set of cervigram images in the IMODT database. This research work obtains 99.22% CPS, 99.03% CPSP, and 99.01% CPA for the set of cervigram images in Guanacaste database.

These experimental results of the proposed work have been significantly compared with the state-of-the-art methods and show the significance and novelty of the proposed works.