BMC Medical Imaging最新文献

In recent years, the incidence of nodular thyroid diseases has been increasing annually. Ultrasonography has become a routine diagnostic tool for thyroid nodules due to its high real-time capabilities and low invasiveness. However, thyroid images obtained from current ultrasound tests often have low resolution and are plagued by significant noise interference. Regional differences in medical conditions and varying levels of physician experience can impact the accuracy and efficiency of diagnostic results. With the advancement of deep learning technology, deep learning models are used to identify whether a nodule in a thyroid ultrasound image is benign or malignant. This helps to close the gap between doctors' experience and equipment differences, improving the accuracy of the initial diagnosis of thyroid nodules. To cope with the problem that thyroid ultrasound images contain complex background and noise as well as poorly defined local features. in this paper, we first construct an improved ResNet50 classification model that uses a two-branch input and incorporates a global attention lightening module. This model is used to improve the accuracy of benign and malignant nodule classification in thyroid ultrasound images and to reduce the computational effort due to the two-branch structure.We constructed a U-net segmentation model incorporating our proposed ACR module, which uses hollow convolution with different dilation rates to capture multi-scale contextual information for feature extraction of nodules in thyroid ultrasound images and uses the results of the segmentation task as an auxiliary branch for the classification task to guide the classification model to focus on the lesion region more efficiently in the case of weak local features. The classification model is guided to focus on the lesion region more efficiently, and the classification and segmentation sub-networks are respectively improved specifically for this study, which is used to improve the accuracy of classifying the benign and malignant nature of the nodules in thyroid ultrasound images. The experimental results show that the four evaluation metrics of accuracy, precision, recall, and f1 of the improved model are 96.01%, 93.3%, 98.8%, and 96.0%, respectively. The improvements were 5.7%, 1.6%, 13.1%, and 7.4%, respectively, compared with the baseline classification model.

Background: Thyroid nodules diagnosed in children pose a greater risk of malignancy compared to those in adults. However, there is no ultrasound thyroid nodule evaluation system aimed at children. The objective of this research is to assess the diagnostic performance of the adult-based American College of Radiology Thyroid Imaging Reporting and Data System (ACR-TIRADS) and Chinese Thyroid Imaging Reporting and Data System (C-TIRADS) in pediatric thyroid carcinoma.

Methods: The preoperative ultrasound images of 177 thyroid lesions in 136 pediatric patients aged 18 or younger who underwent thyroid surgery or fine needle aspiration (FNA) at our center from July 2017 to July 2022 were reviewed. The sonographic characteristics of pediatric thyroid carcinoma were compared and analyzed in contrast to benign nodules. All the nodules were evaluated by the ACR-TIRADS and the C-TIRADS respectively.

Results: Ultrasound features such as solid composition (94.8%), hypoechogenicity or marked hypoechogenicity (94.8-95.7%) and microcalcification (78.3-84.3%) were more common in pediatric malignant nodules (P-values < 0.05). The areas under receiver operating characteristic curves (AUC) of the ACR-TIRADS and the C-TIRADS in diagnosing pediatric thyroid carcinoma were 0.903-0.906, 0.907-0.909 (P-value > 0.05). The interobserver agreement of both the ACR-TIRADS and the C-TIRADS was strong (weighted Kappa > 0.90).

Conclusions: Both the C-TIRADS and the ACR-TIRADS owned great diagnostic performance and strong interobserver agreement in diagnosing pediatric thyroid carcinoma. However, a more complete and specific ultrasound evaluation system for pediatric thyroid nodules is still needed.

Purpose: Systemic immunity is essential for driving therapeutically induced antitumor immune responses, and the spleen may reflect alterations in systemic immunity. This study aimed to evaluate the predictive value of contrast-enhanced CT-based spleen radiomics for progression-free survival (PFS) in patients with locally advanced cervical cancer (LACC) who underwent definitive chemoradiotherapy (dCRT). Additionally, we investigated the role of spleen radiomics features and changes in spleen volume in assessing systemic immunity.

Methods: This retrospective study included 257 patients with LACC who underwent dCRT. The patients were randomly divided into training and validation groups in a 7:3 ratio. Radiomic features were extracted from CT images obtained before and after dCRT. Radiomic scores (Radscore) were calculated using features selected through least absolute shrinkage and selection operator (LASSO) Cox regression. The percentage change in spleen volume was determined from measurements taken before and after treatment. Independent prognostic factors for PFS were identified through multivariate Cox regression analyses. Model performance was evaluated with the receiver operating characteristic (ROC) curve and the C-index. The Radscore cut-off value, determined from the ROC curve, was used to stratify patients into high- and low-risk survival groups. The Wilcoxon test was used to analyze differences in hematological parameters between different survival risk groups and between different spleen volume change groups. Spearman correlation analysis was used to explore the relationship between spleen volume change and hematological parameters.

Results: Independent prognostic factors included FIGO stage, pre-treatment neutrophil-to-lymphocyte ratio (pre-NLR), spleen volume change, and Radscore. The radiomics-combined model demonstrated the best predictive performance for PFS in both the training group (AUC: 0.923, C-index: 0.884) and the validation group (AUC: 0.895, C-index: 0.834). Compared to the low-risk group, the high-risk group had higher pre-NLR (p = 0.0054) and post-NLR (p = 0.038). Additionally, compared to the decreased spleen volume group, the increased spleen volume group had lower post-NLR (p = 0.0059) and post-treatment platelet-to-lymphocyte ratio (p < 0.001).

Conclusion: Spleen radiomics combined with clinical features can effectively predict PFS in patients with LACC after dCRT. Furthermore, spleen radiomics features and changes in spleen volume can reflect alterations in systemic immunity.

Background: T1 mapping can quantify the longitudinal relaxation time of tissues. This study aimed to investigate the repeatability and reproducibility of T1 mapping radiomics features of lung cancer and the feasibility of T1 mapping-based radiomics model to predict its pathological types.

Methods: The chest T1 mapping images and clinical characteristics of 112 lung cancer patients (54 with adenocarcinoma and 58 with other types of lung cancer) were collected retrospectively. 54 patients underwent twice short-term T1 mapping scans. Regions of interest were manually delineated on T1 mapping pseudo-color images to measure the mean native T1 values of lung cancer, and radiomics features were extracted using the semi-automatic segmentation method by two independent observers. The patients were randomly divided into training group (77 cases) and validation group (35 cases) with the ratio of 7:3. Interclass correlation coefficients (ICCs), Student's t-test or Mann-Whitney U tests and least absolute shrinkage and selection operator (LASSO) were used for feature selection. The optimum features were selected to establish a logistic regression (LR) radiomics model. Independent sample t-test, Mann Whitney U-test or chi square test were used to compare the differences of clinical characteristics and T1 values. Performance was compared by the area under the receiver operating characteristic (ROC) curve (AUC).

Results: In the training group, smoking history, lesion type and native T1 values were different between adenocarcinoma and non-adenocarcinoma patients (P = 0.004-0.038). There were 1035 (54.30%) radiomics features meet the intra-and inter-observer, and test-retest reproducibility with ICC > 0.80. After feature dimensionality reduction and model construction, the AUC of T1 mapping-based radiomics model for predicting the pathological types of lung cancer was 0.833 and 0.843, respectively, in the training and validation cohorts. The AUCs of T1 value and clinical model (including smoking history and lesion type) were 0.657 and 0.692 in the training group, and 0.722 and 0.686 in the validation group. Combined with T1 mapping radiomics, clinical model and T1 value to establish a combined model, the prediction efficiency was further improved to 0.895 and 0.915 in the training and validation groups.

Conclusions: About 50% of the T1 mapping-based radiomics features displayed relatively poor repeatability and reproducibility. While T1 mapping-based radiomics model is valuable in identification of histological types of lung cancer despite the measurement variability. Combination of T1 mapping radiomics model, clinical characteristics and native T1 value can improve the predictive value of pathological types of lung cancer.

Background: 3D brachial plexus MRI scanning is prone to examination failure due to the lengthy scan times, which can lead to patient discomfort and motion artifacts. Our purpose is to investigate the efficacy of artificial intelligence-assisted compressed sensing (ACS) in improving the acceleration efficiency and maintaining or enhancing the image quality of brachial plexus MR imaging.

Methods: A total of 30 volunteers underwent 3D sampling perfection with application-optimized contrast using different flip angle evolution short time inversion recovery using a 3.0T MR scanner. The imaging protocol included parallel imaging (PI) and ACS employing acceleration factors of 4.37, 6.22, and 9.03. Radiologists evaluated the neural detail display, fat suppression effectiveness, presence of image artifacts, and overall image quality. Signal intensity and standard deviation of specific anatomical sites within the brachial plexus and background tissues were measured, with signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) subsequently calculated. Cohen's weighted kappa (κ), One-way ANOVA, Kruskal-Wallis and pairwise comparisons with Bonferroni-adjusted significance level. P < 0.05 was considered statistically significant.

Results: ACS significantly reduced scanning times compared to PI. Evaluations revealed differences in subjective scores and SNR across the sequences (P < 0.05), with no marked differences in CNR (P > 0.05). For subjective scores, ACS 9.03 were lower than the other three sequences in neural details display, image artifacts and overall image quality. There was no significant difference in fat suppression. For objective quantitative evaluation, SNR of right C6 root in ACS 6.22 and ACS 9.03 was higher than that in PI; SNR of left C6 root in ACS 4.37, ACS 6.22 and ACS 9.03 was higher than that in PI; SNR of medial cord in ACS 6.22, ACS 9.03 was higher than that in PI.

Conclusion: Compared with PI, ACS can shorten scanning time while ensuring good image quality.

The early detection and diagnosis of gastrointestinal tract diseases, such as ulcerative colitis, polyps, and esophagitis, are crucial for timely treatment. Traditional imaging techniques often rely on manual interpretation, which is subject to variability and may lack precision. Current methodologies leverage conventional deep learning models that, while effective to an extent, often suffer from overfitting and generalization issues on medical image datasets due to the intricate and subtle variations in disease manifestations. These models typically do not fully utilize the potential of transfer learning or advanced data augmentation, leading to less-than-optimal performance, especially in diverse real-world scenarios where data variability is high. This study introduces a robust model using the EfficientNetB5 architecture combined with a sophisticated data augmentation strategy. The model is tailored for the high variability and intricate details present in gastrointestinal tract disease images. By integrating transfer learning with maximal pooling and extensive regularization, the model aims to enhance diagnostic accuracy and reduce overfitting. The proposed model achieved a test accuracy of 98.89%, surpassing traditional methods by incorporating advanced regularization and augmentation techniques. The application of horizontal flipping and dynamic scaling during training significantly improved the model's ability to generalize, evidenced by a low-test loss of 0.230 and high precision metrics across all classes. The proposed deep learning framework demonstrates superior performance in the automated classification of gastrointestinal diseases from image data. By addressing key limitations of existing models through innovative techniques, this study contributes to the enhancement of diagnostic processes in medical imaging, potentially leading to more accurate and timely disease interventions.

Background: Speckle tracking echocardiography (STE) provides quantification of left ventricular (LV) deformation and is useful in the assessment of LV function. STE is increasingly being used clinically, and every effort to simplify and standardize STE is important. Manual outlining of regions of interest (ROIs) is labor intensive and may influence assessment of strain values.

Purpose: We hypothesized that a deep learning (DL) model, trained on clinical echocardiographic exams, can be combined with a readily available echocardiographic analysis software, to automate strain calculation with comparable fidelity to trained cardiologists.

Methods: Data consisted of still frame echocardiographic images with cardiologist-defined ROIs from 672 clinical echocardiographic exams from a university hospital outpatient clinic. Exams included patients with ischemic heart disease, heart failure, valvular disease, and conduction abnormalities, and some healthy subjects. An EfficientNetB1-based architecture was employed, and different techniques and properties including data set size, data quality, augmentations, and transfer learning were evaluated. DL predicted ROIs were reintroduced into commercially available echocardiographic analysis software to automatically calculate strain values.

Results: DL-automated strain calculations had an average absolute difference of 0.75 (95% CI 0.58-0.92) for global longitudinal strain (GLS), and 1.16 (95% CI 1.03-1.29) for single-projection longitudinal strain (LS), compared to operators. A Bland-Altman plot revealed no obvious bias, though there were fewer outliers in the lower average LS ranges. Techniques and data properties yielded no significant increase/decrease in performance.

Conclusion: The study demonstrates that DL-assisted, automated strain measurements are feasible, and provide results within interobserver variation. Employing DL in echocardiographic analyses could further facilitate adoption of STE parameters in clinical practice and research, and improve reproducibility.

Background: To explore the value of computed tomography based radiomics in the differential diagnosis of drug-sensitive and drug-resistant pulmonary tuberculosis.

Methods: The clinical and computed tomography image data of 177 patients who were diagnosed with pulmonary tuberculosis through sputum culture and completed drug-susceptibility testing from April 2018 to December 2020 at the Second Hospital of Nanjing were retrospectively analyzed. Patients with drug-resistant pulmonary tuberculosis (n = 78) and drug-sensitive pulmonary tuberculosis (n = 99) were randomly divided into a training set (n = 124) and a validation set (n = 53) at a ratio of 7:3. Regions of interest were drawn to delineate the lesions and radiomics features were extracted from non-contrast computed tomography images. A radiomics signature based on the valuable radiomics features was constructed and a radiomics score was calculated. Demographic data, clinical symptoms, laboratory results and computed tomography imaging characteristics were evaluated to establish a clinical model. Combined with the Rad-score and clinical factors, a radiomics-clinical model nomogram was constructed.

Results: Thirteen features were used to construct the radiomics signature. The radiomics signature showed good discrimination in the training set (area under the curve (AUC), 0.891; 95% confidence interval (CI), 0.832-0.951) and the validation set (AUC, 0.803; 95% CI, 0.674-0.932). In the clinical model, the AUC of the training set was 0.780(95% CI, 0.700-0.859), while the AUC of the validation set was 0.692 (95% CI, 0.546-0.839). The radiomics-clinical model showed good calibration and discrimination in the training set (AUC, 0.932;95% CI, 0.888-0.977) and the validation set (AUC, 0.841; 95% CI, 0.719-0.962).

Conclusions: Simple radiomics signature is of great value in differentiating drug-sensitive and drug-resistant pulmonary tuberculosis patients. The radiomics-clinical model nomogram showed good predictive, which may help clinicians formulate precise treatments.

Background: To compare the performance of whole tumor and habitats-based computed tomography (CT) radiomics for predicting immunophenotyping in laryngeal squamous cell carcinomas (LSCC) and further evaluate the stratified effect of the radiomics model on disease-free survival (DFS) and overall survival (OS) of LSCC patients.

Methods: In all, 106 LSCC patients (40 with inflamed and 66 with non-inflamed immunophenotyping) were randomly assigned into a training (n = 53) and testing (n = 53) cohort. Briefly, 750 radiomics features from contrast-enhanced CT images were respectively extracted from the whole tumor and two Otsu method-derived subregions. Intraclass correlation coefficients (ICCs) were calculated to evaluate the reproducibility. The radiomics models for predicting immunophenotyping were respectively created using K-nearest neighbors (KNN), logistic regression (LR), and Naive bayes (NB) classifiers. The performance of models in the testing cohort were compared using area under the curve (AUC). The prognostic value of the optimal model was determined by survival analysis.

Results: The radiomics features derived from whole tumor showed better reproducibility than those derived from habitats. The best model for the whole tumor (LR classifier) showed superior performance than that for the habitats (KNN classifier) in the testing cohort, but there were no significant differences (AUC: 0.741 vs. 0.611, p = 0.112). Multivariable Cox regression analysis showed that the immunophenotyping predicted by the optimal model was an independent risk factor of unfavorable DFS (p = 0.009) and OS (p = 0.008) in LSCC patients.

Conclusions: Whole tumor-based CT radiomics could serve as a potential predictive biomarker of immunophenotyping and outcome prediction in LSCC patients.

Background: This study aims to explore the accuracy of Convolutional Neural Network (CNN) models in predicting malignancy in Dynamic Contrast-Enhanced Breast Magnetic Resonance Imaging (DCE-BMRI).

Methods: A total of 273 benign lesions (benign group) and 274 malignant lesions (malignant group) were collected and randomly divided into a training set (246 benign and 245 malignant lesions) and a testing set (28 benign and 28 malignant lesions) in a 9:1 ratio. An additional 53 lesions from 53 patients were designated as the validation set. Five models-VGG16, VGG19, DenseNet201, ResNet50, and MobileNetV2-were evaluated. Model performance was assessed using accuracy (Ac) in the training and testing sets, and precision (Pr), recall (Rc), F1 score (F1), and area under the receiver operating characteristic curve (AUC) in the validation set.

Results: The accuracy of VGG19 on the test set (0.96) is higher than that of VGG16 (0.91), DenseNet201 (0.91), ResNet50 (0.67), and MobileNetV2 (0.88). For the validation set, VGG19 achieved higher performance metrics (Pr 0.75, Rc 0.76, F1 0.73, AUC 0.76) compared to the other models, specifically VGG16 (Pr 0.73, Rc 0.75, F1 0.70, AUC 0.73), DenseNet201 (Pr 0.71, Rc 0.74, F1 0.69, AUC 0.71), ResNet50 (Pr 0.65, Rc 0.68, F1 0.60, AUC 0.65), and MobileNetV2 (Pr 0.73, Rc 0.75, F1 0.71, AUC 0.73). S4 model achieved higher performance metrics (Pr 0.89, Rc 0.88, F1 0.87, AUC 0.89) compared to the other four fine-tuned models, specifically S1 (Pr 0.75, Rc 0.76, F1 0.74, AUC 0.75), S2 (Pr 0.77, Rc 0.79, F1 0.75, AUC 0.77), S3 (Pr 0.76, Rc 0.76, F1 0.73, AUC 0.75), and S5 (Pr 0.77, Rc 0.79, F1 0.75, AUC 0.77). Additionally, S4 model showed the lowest loss value in the testing set. Notably, the AUC of S4 for BI-RADS 3 was 0.90 and for BI-RADS 4 was 0.86, both significantly higher than the 0.65 AUC for BI-RADS 5.

Conclusions: The S4 model we propose has demonstrated superior performance in predicting the likelihood of malignancy in DCE-BMRI, making it a promising candidate for clinical application in patients with breast diseases. However, further validation is essential, highlighting the need for additional data to confirm its efficacy.