BMC Medical Imaging最新文献

Background: This study aimed to perform a comprehensive morphometric analysis of the epiglottis using cone-beam computed tomography (CBCT) images, including the determination of epiglottis dimensions, the investigation of shape variations, and the assessment of their relationship with gender and age.

Methods: A retrospective analysis was conducted on high-quality CBCT images from 100 patients, obtained using the NewTom 5G system. In CBCT images, epiglottis thicknesses (right, midline, left) and horizontal angle at three levels (suprahyoid, hyoid, infrahyoid) were measured in axial sections, while the length and vertical angle of epiglottis were measured in midsagittal view. Epiglottis shapes were identified through 3D visualization.

Results: The midline epiglottis thicknesses were 4.68 mm at the suprahyoid level, 5.51 mm at the hyoid level, and 6.80 mm at the infrahyoid levels. Epiglottis thicknesses and length were statistically significantly greater in males. Of the 100 patients, 51 had a normal curvature, 41 had a flat epiglottis, and 8 had an omega epiglottis. The omega-shaped epiglottis was significantly longer compared to both the flat and normal curvature types (p = 0.011). There was a positive correlation between age and epiglottis thicknesses at the suprahyoid level and horizontal angles at three levels.

Conclusions: This study visualizes epiglottis morphology and uncovers significant morphometric differences. Males exhibit greater epiglottis thickness and length compared to females, while the omega-shaped epiglottis is notably longer than other types. These findings underscore the need for further investigation into the clinical relevance of these morphometric differences, particularly in improving airway management and refining approaches to swallowing function.

Background: A deep learning (DL) model that can automatically detect and classify cervical canal and neural foraminal stenosis using cervical spine magnetic resonance imaging (MRI) can improve diagnostic accuracy and efficiency.

Methods: A method comprising region-of-interest (ROI) detection and cascade prediction was formulated for diagnosing cervical spinal stenosis based on a DL model. First, three part-specific convolutional neural networks were employed to detect the ROIs in different parts of the cervical MR images. Cascade prediction of the stenosis categories was subsequently performed to record the stenosis level and position on each patient slice. Finally, the results were combined to obtain a patient-level diagnostic report. Performance was evaluated based on the accuracy (ACC), area under the curve (AUC), sensitivity, specificity, F1 Score, diagnosis time of the DL model, and recall rate for ROI detection localization.

Results: The average recall rate of the ROI localization was 89.3% (neural foramen) and 99.7% (central canal) under the five-fold cross-validation of the DL model. In the dichotomous classification (normal or mild vs. moderate or severe), the ACC and AUC of the DL model were comparable to those of the radiologists, and the F1 score (84.8%) of the DL model was slightly higher than that of the radiologists (83.8%) for the central canal. Diagnosing whether the central canal or neural foramen of a slice is narrowed in the cervical MRI scan required an average of 15 and 0.098 s for the radiologists and DL model, respectively.

Conclusions: The DL model demonstrated comparable performance with subspecialist radiologists for the detection and classification of central canal and neural foraminal stenosis on cervical spine MRI. Moreover, the DL model demonstrated significant timesaving ability.

Virtual histopathology is an emerging technology in medical imaging that utilizes advanced computational methods to analyze tissue images for more precise disease diagnosis. Traditionally, histopathology relies on manual techniques and expertise, often resulting in time-consuming processes and variability in diagnoses. Virtual histopathology offers a more consistent, and automated approach, employing techniques like machine learning, deep learning, and image processing to simulate staining and enhance tissue analysis. This review explores the strengths, limitations, and clinical applications of these methods, highlighting recent advancements in virtual histopathological approaches. In addition, important areas are identified for future research to improve diagnostic accuracy and efficiency in clinical settings.

Objective: Differentiating intramedullary spinal cord tumor (IMSCT) from spinal cord tumefactive demyelinating lesion (scTDL) remains challenging with standard diagnostic approaches. This study aims to develop and evaluate the effectiveness of a magnetic resonance imaging (MRI)-based radiomics model for distinguishing scTDL from IMSCT before treatment initiation.

Methods: A total of 75 patients were analyzed in this retrospective study, comprising 55 with IMSCT and 20 with scTDL. Radiomics features were extracted from T1- and T2-weighted imaging (T1&T2WI) scans upon admission. Ten classification algorithms were employed: logistic regression (LR); naive bayes (NaiveBayes); support vector machine (SVM); k nearest neighbors (KNN); random forest (RF); extra trees (ExtraTrees); eXtreme gradient boosting (XGBoost); light gradient boosting machine (LightGBM); gradient boosting (GradientBoosting); and multi-Layer perceptron (MLP). The performance of the optimal model was then compared to radiologists' assessments.

Results: This study developed 30 predictive models using ten classifiers across two imaging sequences. The MLP model with two sequences (T1&T2WI) emerged as the most effective one, showing superior accuracy in MRI analysis with an area under the curve (AUC) of 0.991 in training and 0.962 in testing. Moreover, statistical analyses highlighted the radiomics model significantly outperformed radiologists' assessments (p < 0.05) in distinguishing between IMSCT and scTDL.

Conclusion: We present an MRI-based radiomics model with high diagnostic accuracy in differentiating IMSCT from scTDL. The model's performance was comparable to junior radiologists, highlighting its potential as an effective diagnostic aid in clinical practice.

Introduction: The treatment of preinvasive lesions is more effective than treating invasive disease, hence detecting cancer at its early stages is crucial. However, currently, available screening methods show various limitations in terms of sensitivity, specificity, and practicality, thus novel markers complementing traditional cyto/histopathological assessments are needed. Alteration in choline metabolism is a hallmark of many malignancies, including cervical and breast cancers. Choline radiotracers are widely used for imaging purposes, even though many risks are associated with their radioactivity. Therefore, this work aimed to synthesise and characterise a non-radioactive choline tracer based on a fluorinated acridine scaffold (CFA) for the in vitro detection of cervical and breast cancer cells by fluorescence imaging.

Methods: CFA was fully characterised and tested for its cytotoxicity on breast (MCF-7), cervical (HeLa), glioblastoma (U-87 MG) and hepatoblastoma (HepG2) cancer cell lines and in normal cell lines (epithelial, HEK-293 and human dermal fibroblasts, HDFs). The cellular uptake of CFA was investigated by a confocal microscope and its accumulation was quantified over time. The specificity of CFA over mesenchymal origin cells (HDFs), as a model of cancer-associated fibroblasts was investigated by fluorescence microscopy.

Results: CFA was toxic at much higher concentrations (HeLa IC₅₀ = 200 ± 18 µM and MCF-7 IC₅₀ = 105 ± 3 µM) than needed for its detection in cancer cells (5 µM). CFA was not toxic in the other cell lines tested. The intensity of CFA in breast and cervical cancer cells was not significantly different at any time point, yet it was greater than HepG2 and U-87 MG (p ≤ 0.01 and p ≤ 0.0001, respectively) after 24 h incubation. A very weak signal intensity was recorded in HEK-293 and HDFs (p ≤ 0.001 and p ≤ 0.0001, respectively). A selective ability of CFA to accumulate in HeLa and MCF-7 was recorded upon co-culture with fibroblasts.

Conclusions: The results showed that CFA preferentially accumulated in cancer cells rather than in normal cells. These findings suggest that CFA may be a potential diagnostic probe for discriminating healthy tissues from malignant tissues due to its specific and highly sensitive features; CFA may also represent a useful tool for in vitro/ex vivo investigations of choline metabolism in patients with cervical and breast cancers.

Background and purpose: Tumor bed (TB) is the residual cavity of resected tumor after surgery. Delineating TB from CT is crucial in generating clinical target volume for radiotherapy. Due to multiple surgical effects and low image contrast, segmenting TB from soft tissue is challenging. In clinical practice, titanium clips were used as marks to guide the searching of TB. However, this information is limited and may cause large error. To provide more prior location information, the tumor regions on both pre-operative and post-operative CTs are both used by the deep learning model in segmenting TB from surrounding tissues.

Materials and methods: For breast cancer patient after surgery and going to be treated by radiotherapy, it is important to delineate the target volume for treatment planning. In clinical practice, the target volume is usually generated from TB by adding a certain margin. Therefore, it is crucial to identify TB from soft tissue. To facilitate this process, a deep learning model is developed to segment TB from CT with the guidance of prior tumor location. Initially, the tumor contour on the pre-operative CT is delineated by physician for surgical planning purpose. Then this contour is transformed to the post-operative CT via the deformable image registration between paired pre-operative and post-operative CTs. The original and transformed tumor regions are both used as inputs for predicting the possible region of TB by the deep-learning model.

Results: Compared to the one without prior tumor contour information, the dice similarity coefficient of the deep-learning model with the prior tumor contour information is improved significantly (0.812 vs. 0.520, P = 0.001). Compared to the traditional gray-level thresholding method, the dice similarity coefficient of the deep-learning model with the prior tumor contour information is improved significantly (0.812 vs.0.633, P = 0.0005).

Conclusions: The prior tumor contours on both pre-operative and post-operative CTs provide valuable information in searching for the precise location of TB on post-operative CT. The proposed method provided a feasible way to assist auto-segmentation of TB in treatment planning of radiotherapy after breast-conserving surgery.

Background: Esophageal fistula (EF), a rare and potentially fatal complication, can be better managed with predictive models for personalized treatment plans in esophageal cancers. We aim to develop a clinical-deep learning radiomics model for effectively predicting the occurrence of EF.

Methods: The study involved esophageal cancer patients undergoing radiotherapy or chemoradiotherapy. Arterial phase enhanced CT images were used to extract handcrafted and deep learning radiomic features. Along with clinical information, a 3-step feature selection method (statistical tests, Least Absolute Shrinkage and Selection Operator, and Recursive Feature Elimination) was used to identify five feature sets in training cohort for constructing random forest EF prediction models. Model performance was compared and validated in both retrospective and prospective test cohorts.

Results: One hundred seventy five patients (122 in training and 53 in test cohort)were retrospectively collected from April 2018 to June 2022. An additional 27 patients were enrolled as a prospective test cohort from June 2022 to December 2023. Post-selection in the training cohort, five feature sets were used for model construction: clinical, handcrafted radiomic, deep learning radiomic, clinical-handcrafted radiomic, and clinical-deep learning radiomic. The clinical-deep learning radiomic model excelled with AUC of 0.89 (95% Confidence Interval: 0.83-0.95) in the training cohort, 0.81 (0.65-0.94) in the test cohort, and 0.85 (0.71-0.97) in the prospective test cohort. Brier-score and calibration curve analyses validated its predictive ability.

Conclusions: The clinical-deep learning radiomic model can effectively predict EF in patients with advanced esophageal cancer undergoing radiotherapy or chemoradiotherapy.

Background: Ovarian cancer remains a leading cause of death among women, largely due to its asymptomatic early stages and high mortality when diagnosed late. Early detection significantly improves survival rates, and the Ovarian-Adnexal Reporting and Data System Ultrasound (O-RADS US) is currently the most commonly used method, but has limitations in specificity and accuracy. While O-RADS US has standardized reporting, its sensitivity can lead to the misdiagnosis of benign masses as malignant, resulting in overtreatment. This study aimed to construct a nomogram model based on the O-RADS US and clinical and laboratory indicators to predict the malignancy risk of adnexal cystic-solid masses.

Methods: This retrospective study collected data from patients with adnexal cystic-solid masses who underwent ultrasonography and were pathologically confirmed between January 2021 and December 2023 at the First Affiliated Hospital of Shenzhen University. They were categorized into benign and malignant groups according to pathological findings. Least absolute shrinkage and selection operator (LASSO) regression analysis was used to select the most relevant predictors of ovarian cancer. A nomogram model was constructed, and its diagnostic performance was calculated. We bootstrapped the data 500 times to perform internal verification, drew a calibration curve to verify the prediction ability, and performed a decision curve analysis to assess clinical usefulness.

Results: A total of 399 patients with adnexal cystic-solid masses were included in this study: 327 in the benign group and 72 in the malignant group. Five predictors associated with the risk of malignancy of adnexal cystic-solid masses were selected using LASSO regression: O-RADS, acoustic shadowing, postmenopausal status, CA125, and HE4. The area under the curve, sensitivity, specificity, accuracy, positive and negative predictive values of the nomogram were 0.909, 83.3%, 82.9%, 83.0%, 51.7%, and 95.8%, respectively. The calibration curve of the nomogram showed good consistency between the predicted and actual probabilities, and the decision curve showed good clinical usefulness.

Conclusion: The nomogram model based on O-RADS US and clinical and laboratory indicators can be used to predict the risk of malignancy in adnexal cystic-solid masses, with high predictive performance, good calibration, and clinical usefulness.

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.