BMC Medical Imaging最新文献

Objective: This study was to develop a multi-parametric MRI radiomics model to predict preoperative Ki-67 status.

Materials and methods: A total of 120 patients with pathologically confirmed breast cancer were retrospectively enrolled and randomly divided into a training set (n = 84) and a validation set (n = 36). Radiomic features were derived from both the intratumoral and peritumoral regions, extending 5 mm from the tumor boundary, using magnetic resonance imaging (MRI). The MRI sequences employed included T2-weighted imaging (T2WI), dynamic contrast-enhanced (DCE) imaging, diffusion-weighted imaging (DWI), and apparent diffusion coefficient (ADC) maps. The T-test and the Least Absolute Shrinkage and Selection Operator Cross-Validation (LASSO CV) were conducted for feature selection. Model_intra, model_peri, model_intra+peri were established by eleven supervised machine learning (ML) algorithms to predict the expression status of Ki-67 in breast cancer and were verified by the validation groups. The model's performance was evaluated by employing metrics such as the area under the curve (AUC), accuracy, sensitivity, and specificity.

Results: The features of intratumor, peritumor, intratumor + peritumor were extracted 851, 851 and 1702 samples respectively, 14, 23 and 35 features were selected by LASSO. ML algorithms based on model_intra and model_peri consistently yield AUCs that are below 80% in the validation set. Hower, Logistic regression (LR) and linear discriminant analysis (LDA) based on model_intra+peri demonstrated significant advantages over other algorithms, achieving AUCs of 0.92 and 0.98, accuracies of 0.94 and 0.97, sensitivities of 1 and 0.96, and specificities of 0.85 and 1 respectively in the validation set.

Conclusion: The integrated intra- and peritumoral radiomics model, developed using multiparametric MRI data and machine learning classifiers, exhibits significant predictive power for Ki-67 expression levels. This model could facilitate personalized clinical treatment strategies for individuals diagnosed with breast cancer (BC).

Clinical trial number: Not applicable.

Objective: The study aimed to evaluate the application value of computed tomography (CT) three-dimensional (3D) reconstruction technology in identifying benign and malignant lung nodules and characterizing the distribution of the nodules.

Methods: CT 3D reconstruction was performed for lung nodules. Pathological results were used as the gold standard to compare the detection rates of various lung nodule signs between conventional chest CT scanning and CT 3D reconstruction techniques. Additionally, the differences in mean diffusion coefficient values and partial anisotropy index values between male and female patients were analyzed.

Results: Pathologic confirmation identified 30 patients with benign lesions and 45 patients with malignant lesions. CT 3D reconstruction demonstrated higher diagnostic accuracy for lung nodule imaging signs compared to conventional CT scanning (P < 0.05). The mean diffusion coefficient values and partial anisotropy index values were lower in female patients compared to male patients in the lung nodule lesion area, lung perinodular edema area, and normal lung tissue (P < 0.05). Conventional CT scanning showed a benign accuracy rate of 63.33% and a malignant accuracy rate of 60.00%, whereas CT 3D imaging achieved a benign and malignant accuracy rate of 86.67% for both. The accuracy rates for CT 3D imaging were significantly higher than those for conventional CT scanning (P < 0.05).

Conclusion: CT 3D imaging technology demonstrates high diagnostic accuracy in differentiating benign from malignant lung nodules.

Invasive breast cancer diagnosis and treatment planning require an accurate assessment of human epidermal growth factor receptor 2 (HER2) expression levels. While immunohistochemical techniques (IHC) are the gold standard for HER2 evaluation, their implementation can be resource-intensive and costly. To reduce these obstacles and expedite the procedure, we present an efficient deep-learning model that generates high-quality IHC-stained images directly from Hematoxylin and Eosin (H&E) stained images. We propose a new IHC-GAN that enhances the Pix2PixHD model into a dual generator module, improving its performance and simplifying its structure. Furthermore, to strengthen feature extraction for HE-stained image classification, we integrate MobileNetV3 as the backbone network. The extracted features are then merged with those generated by the generator to improve overall performance. Moreover, the decoder's performance is enhanced by providing the related features from the classified labels by incorporating the adaptive instance normalization technique. The proposed IHC-GAN was trained and validated on a comprehensive dataset comprising 4,870 registered image pairs, encompassing a spectrum of HER2 expression levels. Our findings demonstrate promising results in translating H&E images to IHC-equivalent representations, offering a potential solution to reduce the costs associated with traditional HER2 assessment methods. We extensively validate our model and the current dataset. We compare it with state-of-the-art techniques, achieving high performance using different evaluation metrics, showing 0.0927 FID, 22.87 PSNR, and 0.3735 SSIM. The proposed approach exhibits significant enhancements over current GAN models, including an 88% reduction in Frechet Inception Distance (FID), a 4% enhancement in Learned Perceptual Image Patch Similarity (LPIPS), a 10% increase in Peak Signal-to-Noise Ratio (PSNR), and a 45% reduction in Mean Squared Error (MSE). This advancement holds significant potential for enhancing efficiency, reducing manpower requirements, and facilitating timely treatment decisions in breast cancer care.

A bone fracture is a medical condition characterized by a partial or complete break in the continuity of the bone. Fractures are primarily caused by injuries and accidents, affecting millions of people worldwide. The healing process for a fracture can take anywhere from one month to one year, leading to significant economic and psychological challenges for patients. The detection of bone fractures is crucial, and radiographic images are often relied on for accurate assessment. An efficient neural network method is essential for the early detection and timely treatment of fractures. In this study, we propose a novel transfer learning-based approach called MobLG-Net for feature engineering purposes. Initially, the spatial features are extracted from bone X-ray images using a transfer model, MobileNet, and then input into a tree-based light gradient boosting machine (LGBM) model for the generation of class probability features. Several machine learning (ML) techniques are applied to the subsets of newly generated transfer features to compare the results. K-nearest neighbor (KNN), LGBM, logistic regression (LR), and random forest (RF) are implemented using the novel features with optimized hyperparameters. The LGBM and LR models trained on proposed MobLG-Net (MobileNet-LGBM) based features outperformed others, achieving an accuracy of 99% in predicting bone fractures. A cross-validation mechanism is used to evaluate the performance of each model. The proposed study can improve the detection of bone fractures using X-ray images.

Background: This retrospective study aims to characterise the root canal morphology of maxillary and mandibular second molars using cone-beam computed tomography (CBCT). The number of roots and canal configurations were evaluated using both the Vertucci and Benjamı´n Brisen˜ o Marroquı´n classification systems.

Methods: A total of 1084 second molar images (523 maxillary; 266 right and 257 left side and 561 mandibular; 285 right and 276 left side) were evaluated from 320 CBCT scans analyzed for the Turkish subpopulation. CBCT imaging provided superior visualisation of root canal anatomy compared to periapical radiography. The findings revealed diverse root canal configurations, with variations observed even within the same population. Statistical analyses, including the chi-squared test, were used to assess correlations between root number and demographic variables such as age and sex.

Results: According to Benjamı´n Brisen˜ o Marroquı´n classification system, the most common configuration for upper right three-rooted teeth mesial root was ³URM^2-1 (n:66, 35.7%), for distal root was ³URM¹ (n:169, 91.4%), and for palatal root was ³URM¹ (n:165, 89.2%). Additionally, the most common configuration for upper left three-rooted teeth mesial root was ³27¹ (n:50, 28.4%), for distal root was ³ULM¹ (n:160, 90.9%), and for palatal root was ³ULM¹ (n:158, 89.8%). In lower left molars, the most common configuration in the two-rooted teeth mesial root was ²LLM² (n:114, 49.4%), and for the distal root was ²LLM¹ (n:170, 73.6%). For lower right the most common configuration for two-rooted teeth mesial root was ²LRM² (n:125, 52.5%), and for distal root was ²LRM¹ (n:173, 72.7%)(p < 0.05).

Conclusion: The primary outcome was observed that the root canal anatomy of upper and lower second molars may differ in both classifications of Turkish subpopulation. While Vertucci's classification was inadequate in some cases, Briseno-Marroquin classification was able to classify all upper and lower second molars with a single code. This new classification is a more useful system for classifying all second molars. There is a statistically significant difference exists among the new configuration according to the distribution of the teeth analyzed.

Background: This prospective study tested the hypothesis that the apparent diffusion coefficient (ADC) value and tumor volume (TV) measured in diffusion-weighted magnetic resonance imaging (DW-MRI) before, during, and after the treatment are quantitative imaging markers to assess tumor response in muscle-invasive bladder cancer (MIBC) patients undergoing neoadjuvant chemotherapy (NAC).

Methods: Multi-parametric MRI was prospectively done for MIBC patients at 3 time points. Pre-treatment ADC value, pre-treatment TV, as well as, percent of changes (ΔADC%, and ΔTV%) in these parameters at mid- and post-treatment relative to baseline were calculated and compared between the patients with and without clinical complete response (CR). Also, further analysis was carried out based on the groups of patients with and without overall response (OR). Two different methods of ADC estimation including single-slice ADC measurement (ADC_single-slice) and whole-lesion ADC measurement (ADC_whole-lesion) were used.

Results: A total of 50 eligible patients were included in the analysis. Of these, 20 patients (40%) showed clinical CR to treatment, while 30 (60%) did not. Our results showed that although there was no significant difference between the two groups of patients with and without CR in terms of mid-treatment ΔADC% and mid-treatment ΔTV%, significant differences were observed in terms of the pre-treatment ADC (p < 0.01), pre-treatment TV (p < 0.001), post-treatment ΔADC% (p < 0.05), and post-treatment ΔTV% (p < 0.05). The results of the OR-based analysis were in line with the CR-based results. There was also a strong and significant correlation between ADC_single-slice and ADC_whole-lesion measurements (r > 0.9, P < 0.001).

Conclusion: Pre-treatment ADC, pre-treatment TV, post-treatment ΔADC%, and post-treatment ΔTV% could be considered as promising quantitative imaging markers of tumor response in MIBC patients undergoing NAC. Moreover, mid-treatment ΔADC% and mid-treatment ΔTV% should not be used as predictors of tumor response in these patients. Further larger studies are required to confirm these results.

Objectives: To develop ultrasound-based radiomics models and a clinical model associated with inflammatory markers for predicting intrahepatic cholangiocarcinoma (ICC) lymph node (LN) metastasis. Both are integrated for enhanced preoperative prediction.

Methods: This study retrospectively enrolled 156 surgically diagnosed ICC patients. A region of interest (ROI) was manually identified on the ultrasound image of the tumor to extract radiomics features. In the training cohort, we performed a Wilcoxon test to screen for differentially expressed features, and then we used 12 machine learning algorithms to develop 107 models within the cross-validation framework and determine the optimal radiomics model through receiver operating characteristic (ROC) curve analysis. Multivariable logistic regression analysis was used to identify independent risk factors to construct a clinical model. The combined model was established by combining ultrasound-based radiomics and clinical parameters. The Delong test and decision curve analysis (DCA) were used to compare the diagnostic efficacy and clinical utility of different models.

Results: A total of 1239 radiomics features were extracted from the ROIs of tumors. Among the 107 prediction models, the model (Stepglm + LASSO) utilizing 10 radiomics features ultimately yielded the highest average area under the receiver operating characteristic curve (AUC) of 0.872, with an AUC of 0.916 in the training cohort and 0.827 in the validation cohort. The combined model, which incorporates the optimal radiomics score, clinical N stage, and platelet-to-lymphocyte ratio (PLR), achieved an AUC of 0.882 in the validation cohort, significantly outperforming the clinical model with an AUC of 0.687 (P = 0.009). According to the DCA analysis, the combined model also showed better clinical benefits.

Conclusions: The combined model incorporating ultrasound-based radiomics features and the PLR marker offers an effective, noninvasive intelligence-assisted tool for preoperative LN metastasis prediction in ICC patients.

Clinical trial number: Not applicable.

Purpose: We aimed to perform a Bayesian network meta-analysis to assess the comparative diagnostic performance of different imaging modalities in chronic pancreatitis(CP).

Methods: The PubMed, Embase and Cochrane Library databases were searched for relevant publications until March 2024. All studies evaluating the head-to-head diagnostic performance of imaging modalities in CP were included. Bayesian network meta-analysis was performed to compare the sensitivity and specificity between the imaging modalities. The Quality Assessment of Diagnostic Performance Studies (QUADAS-2) tool was used to evaluate the quality of studies.

Results: This meta-analysis incorporated 17 studies. Network meta-analytic results indicated that endoscopic ultrasonography (EUS) achieved the highest surface under the cumulative ranking (SUCRA) value at 0.86 for sensitivity. Conversely, magnetic resonance imaging (MRI) demonstrated best specificity, recording the highest SUCRA value at 0.99. Ultrasonography (US) displayed comparatively lower sensitivity than endoscopic retrograde cholangiopancreatography (ERCP) (relative risk[RR]: 0.83, 95% Confidence Interval[CI]: 0.69-0.99) and EUS (RR: 0.73, 95% CI: 0.57-0.91). MRI outperformed all other imaging modalities in terms of specificity.

Conclusions: It appears that EUS demonstrates higher sensitivity, while MRI exhibits higher specificity in patients with chronic pancreatitis. However, it is crucial to note that our analysis was limited to the diagnostic performance and did not evaluate the cost-effectiveness of these various imaging modalities. Consequently, further extensive studies are needed to assess the benefit-to-risk ratios comprehensively.

Objective: This study investigates the MRI characteristics of primary and metastatic adult granulosa cell tumor with normal estrogen levels (AGCT-NEL) to enhance clinical understanding and diagnostic accuracy of this disease.

Methods: We collected clinical data from 10 patients with AGCT-NEL, confirmed by pathology, treated at our hospital from January 2016 to January 2024. We retrospectively analyzed the MRI features of primary and metastatic lesions from aspects such as shape, edge characteristics, MRI signal, and enhancement features.

Results: A total of 10 AGCT-NEL patients were included in this study, aged 28 to 81 years, with an average age of 54 ± 16 years. The primary tumors primarily presented as unilocular cystic, solid, and cystic-solid types. The solid components showed isointense to slightly hypointense signals on T1-weighted imaging (T₁WI), slightly hyperintense signals on T2-weighted imaging (T₂WI), and high signals on diffusion-weighted imaging (DWI), with possible internal hemorrhage or cystic degeneration. The cystic components exhibited low signal on T₁WI, high signal on T₂WI, uniform wall thickness, and no wall nodules, typically showing hemorrhagic fluid levels. Honeycomb and Swiss cheese signs are sometimes observed in cystic-solid tumors. All metastatic lesions were cystic (either unilocular or multilocular), presenting low signal on T₁WI and high signal on T₂WI, with no wall nodules and possible internal hemorrhagic fluid levels. The multilocular metastatic tumors demonstrated unevenly thickened partitions, also displaying honeycomb and Swiss cheese signs.

Conclusion: The MRI characteristics of primary and metastatic lesions in AGCT-NEL possess specific features, such as signs of hemorrhage, absence of wall nodules in the cystic portions of the tumors, and distinctive honeycomb and Swiss cheese signs, with metastatic lesions being cystic. Understanding these features can aid in improving preoperative diagnostic capabilities and reducing misdiagnosis.

Background: To develop and validate an interpretable machine learning model based on intratumoral and peritumoral radiomics combined with clinicoradiological features and metabolic information from magnetic resonance spectroscopy (MRS), to predict clinically significant prostate cancer (csPCa, Gleason score ≥ 3 + 4) and avoid unnecessary biopsies.

Methods: This study retrospectively analyzed 350 patients with suspicious prostate lesions from our institution who underwent 3.0 Tesla multiparametric magnetic resonance imaging (mpMRI) prior to biopsy (training set, n = 191, testing set, n = 83, and a temporal validation set, n = 76). Intratumoral and peritumoral volumes of interest (VOI_intra, VOI_peri)) were manually segmented by experienced radiologists on T2-weighted imaging (T2WI) and apparent diffusion coefficient (ADC) maps. Radiomic features were extracted separately from the VOI_intra and VOI_peri. After feature selection via the recursive feature elimination (RFE) algorithm, intratumoral radiomic score (intra-rad-score) and peritumoral radiomic score (peri-rad-score) were constructed. The clinical model, MRS model, and combined model integrating radiomic, clinicoradiological and metabolic features were constructed via the eXtreme Gradient Boosting (XGBoost) algorithm. The predictive performance of the models was evaluated in both the training and testing sets using receiver operating characteristic (ROC) curve analysis. SHapley Additive exPlanations (SHAP) analysis was applied to the combined model to visualize and interpret the prediction process.

Results: A total of 350 patients were included, comprising 173 patients with csPCa (49.4%) and 177 patients with non-csPCa (50.6%). The intra-rad-score and peri-rad-score were constructed via 10 and 16 radiomic features. The combined model demonstrated the highest AUC, accuracy, F1 score, sensitivity, and specificity in the testing set (0.968, 0.928, 0.927, 0.932, and 0.923, respectively) and in the temporal validation set (0.940, 0.895, 0.890, 0.923, and 0.875, respectively). SHAP analysis revealed that the intra-rad-score, PSAD, peri-rad-score, and PI-RADS score were the most important predictors of the combined model.

Conclusion: We developed and validated a robust machine learning model incorporating intratumoral and peritumoral radiomic features, along with clinicoradiological and metabolic parameters, to accurately identify csPCa. The prediction process was visualized via SHAP analysis to facilitate clinical decision- making.