Current Medical Imaging Reviews最新文献

Introduction: Colorectal cancer (CRC) is a leading cause of cancer-related morbidity and mortality. Accurate differentiation of tumor grade is crucial for prognosis and treatment planning. This study aimed to evaluate the diagnostic value of dual-source CT dual-energy technology parameters in distinguishing CRC differentiation grades.

Methods: A retrospective analysis was conducted on 87 surgically and pathologically confirmed CRC patients (64 with medium-high differentiation and 23 with low differentiation) who underwent dual-source CT dual-energy enhancement scanning. Normalized iodine concentration (NIC), spectral curve slope (K), and dual-energy index (DEI) of the tumor center were measured in arterial and venous phases. Differences in these parameters between differentiation groups were compared, and ROC curve analysis was performed to assess diagnostic efficacy.

Results: The low-differentiation group exhibited significantly higher NIC, K, and DEI values in both arterial and venous phases compared to the mediumhigh differentiation group (P < 0.01). In the arterial phase, NIC, K, and DEI yielded AUC values of 0.920, 0.770, and 0.903, respectively, with sensitivities of 95.7%, 65.2%, and 91.3%, and specificities of 82.8%, 75.0%, and 75.0%, respectively. In the venous phase, AUC values were 0.874, 0.837, and 0.886, with sensitivities of 91.3%, 82.6%, and 91.3%, and specificities of 68.75%, 75.0%, and 73.4%. NIC in the arterial phase showed statistically superior diagnostic performance compared to K values (P < 0.05).

Discussion: Dual-energy CT parameters, particularly NIC in the arterial phase, demonstrate high diagnostic accuracy in differentiating CRC grades. These findings suggest that quantitative dual-energy CT metrics can serve as valuable non-invasive tools for tumor characterization, aiding in clinical decision-making. Study limitations include its retrospective design and relatively small sample size.

Conclusion: NIC, K, and DEI values in dual-energy CT scans are highly effective in distinguishing CRC differentiation grades, with arterial-phase NIC showing the highest diagnostic performance. These parameters may enhance preoperative assessment and personalized treatment strategies for CRC patients.

Introduction: This study explored a generative image synthesis method based on diffusion models, potentially providing a low-cost and high-efficiency training data augmentation strategy for medical artificial intelligence (AI) applications.

Methods: The MedMNIST v2 dataset was utilized as a small-volume training dataset under low-performance computing conditions. Based on the characteristics of existing samples, new medical images were synthesized using the proposed annotated diffusion model. In addition to observational assessment, quantitative evaluation was performed based on the gradient descent of the loss function during the generation process and the Fréchet Inception Distance (FID), using various loss functions and feature vector dimensions.

Results: Compared to the original data, the proposed diffusion model successfully generated medical images of similar styles but with dramatically varied anatomic details. The model trained with the Huber loss function achieved a higher FID of 15.2 at a feature vector dimension of 2048, compared with the model trained with the L2 loss function, which achieved the best FID of 0.85 at a feature vector dimension of 64.

Discussion: The use of the Huber loss enhanced model robustness, while FID values indicated acceptable similarity between generated and real images. Future work should explore the application of these models to more complex datasets and clinical scenarios.

Conclusion: This study demonstrated that diffusion model-based medical image synthesis is potentially applicable as an augmentation strategy for AI, particularly in situations where access to real clinical data is limited. Optimal training parameters were also proposed by evaluating the dimensionality of feature vectors in FID calculations and the complexity of loss functions.

Introduction: Accurate liver volumetry is crucial for hepatectomy. In this study, we developed and validated a deep learning system for automated liver volumetry in patients undergoing hepatectomy, both preoperatively and at 7 days and 3 months postoperatively.

Methods: A 3D U-Net model was trained on CT images from three time points using a five-fold cross-validation approach. Model performance was assessed with standard metrics and comparatively evaluated across the time points.

Results: The model achieved a mean Dice Similarity Coefficient (DSC) of 94.31% (preoperative: 94.91%; 7-day post-operative: 93.45%; 3-month postoperative: 94.57%) and a mean recall of 96.04%. The volumetric difference between predicted and actual volumes was 1.01 ± 0.06% preoperatively, compared to 1.04 ± 0.03% at other time points (p < 0.05).

Discussion: This study demonstrates a novel capability to automatically track post-hepatectomy regeneration using AI, offering significant potential to enhance surgical planning and patient monitoring. A key limitation, however, was that the direct correlation with clinical outcomes was not assessed due to constraints of the current dataset. Therefore, future studies using larger, multi-center datasets are essential to validate the model's clinical and prognostic utility.

Conclusion: The developed artificial intelligence model successfully and accurately measured liver volumes across three critical post-hepatectomy time points. These findings support the use of this automated technology as a precise and reliable tool to assist in surgical decision-making and postoperative assessment, providing a strong foundation for enhancing patient care.

Introduction: Anemia is a common blood disorder caused by a low red blood cell count, reducing blood hemoglobin. It affects children, adolescents, and adults of all genders. Anemia diagnosis typically involves invasive procedures like peripheral blood smears and complete blood count (CBC) analysis. This study aims to develop a cost-effective, non-invasive tool for anemia detection using eye conjunctiva images.

Method: Eye conjunctiva images were captured from 54 subjects using three imaging modalities such as a DSLR camera, a smartphone camera, and a smartphone camera fitted with a 3D-printed spacer macro lens. Image processing techniques, including You Only Look Once (YOLOv8) and the Segment Anything Model (SAM), and K-means clustering were used to analyze the image. By using an MLP classifier, the images were classified as anemic, moderately anemic, and normal. The trained model was embedded into an Android application with geotagging capabilities to map the prevalence of anemia in different regions.

Results: Features extracted using SAM segmentation showed higher statistical significance (p < 0.05) compared to K-Means. Comparing high resolution(DSLR modality) and the proposed 3D-printed spacer macrolens shows statistically significant differences (p < 0.05). The classification accuracy was 98.3% for images from a 3D spacer-equipped smartphone camera, on par with the 98.8% accuracy obtained from DSLR camerabased images.

Conclusion: The mobile application, developed using images captured with a 3D spacer-equipped modality, provides portable, cost-effective, and user-friendly non-invasive anemia screening. By identifying anemic clusters, it assists healthcare workers in targeted interventions and supports global health initiatives like Sustainable Development Goal (SDG) 3.

Introduction: Accurate preoperative grading of adult-type diffuse gliomas is crucial for personalized treatment. Emerging evidence suggests tumor cell infiltration extends beyond peritumoral edema, but the predictive value of radiomics features in these regions remains underexplored.

Method: A retrospective analysis was conducted on 180 patients from the UCSF-PDGM dataset, split into training (70%) and validation (30%) cohorts. Intratumoral volumes (VOI_I, including tumor body and edema) and peritumoral volumes (VOI_P) at 7 expansion distances (1-5, 10, 15 mm) were analyzed. Feature selection involved Levene's test, t-test, mRMR, and LASSO regression. Radiomics models (VOI_I, VOI_P, and combined intratumoral-peritumoral models) were evaluated using AUC, accuracy, sensitivity, specificity, and F1 score, with Delong tests for comparisons.

Results: The combined radiomics models established for the intratumoral and peritumoral 1-5mm ranges (VOI_1-5mm) showed better predictive performance than the VOI_I model (AUC=0.815/0.672), among which the VOI_1 model performed the best: in the training cohort, the AUC was 0.903 (accuracy=0.880, sensitivity=0.905, specificity=0.855, F1=0.884); in the validation cohort, the AUC was 0.904 (accuracy=0.852, sensitivity=0.778, specificity=0.926, F1=0.840). This model significantly outperformed the VOI_I model (p<0.05) and the 10/15mm combined models (p<0.05).

Discussion: The peritumoral regions within 5 mm beyond the edematous area contain critical grading information, likely reflecting subtle tumor infiltration. Model performance declined with larger peritumoral distances, possibly due to increased normal tissue dilution.

Conclusion: The radiomics features of the intratumoral region and the peritumoral region within 5 mm can optimize the preoperative grading of gliomas, providing support for surgical planning and prognostic evaluation.

Introduction: Lung cancer remains a leading cause of cancer-related mortality worldwide. While deep learning approaches show promise in medical imaging, comprehensive comparisons of classifier combinations with DenseNet architectures for lung cancer classification are limited. The study investigates the performance of different classifier combinations, Support Vector Machine (SVM), Artificial Neural Network (ANN), and Multi-Layer Perceptron (MLP), with DenseNet architectures for lung cancer classification using chest CT scan images.

Methods: A comparative analysis was conducted on 1,000 chest CT scan images comprising Adenocarcinoma, Large Cell Carcinoma, Squamous Cell Carcinoma, and normal tissue samples. Three DenseNet variants (DenseNet-121, DenseNet-169, DenseNet-201) were combined with three classifiers: SVM, ANN, and MLP. Performance was evaluated using accuracy, Area Under the Curve (AUC), precision, recall, specificity, and F1- score with an 80-20 train-test split.

Results: The optimal model achieved 92% training accuracy and 83% test accuracy. Performance across models ranged from 81% to 92% for training accuracy and 73% to 83% for test accuracy. The most balanced combination demonstrated robust results (training: 85% accuracy, 0.99 AUC; test: 79% accuracy, 0.95 AUC) with minimal overfitting.

Discussion: Deep learning approaches effectively categorize chest CT scans for lung cancer detection. The MLP-DenseNet-169 combination's 83% test accuracy represents a promising benchmark. Limitations include retrospective design and a limited sample size from a single source.

Conclusion: This evaluation demonstrates the effectiveness of combining DenseNet architectures with different classifiers for lung cancer CT classification. The MLP-DenseNet-169 achieved optimal performance, while SVM-DenseNet-169 showed superior stability, providing valuable benchmarks for automated lung cancer detection systems.

Background: Microsurgical treatment of posterior circulation aneurysms remains challenging due to their deep location, complex anatomical exposure, and close proximity to critical neurovascular structures. Ensuring adequate collateral circulation is paramount for preventing ischemic complications. Indocyanine Green (ICG) and Fluorescein Video Angiography (FL-VAG) have emerged as effective intraoperative tools for assessing cerebral perfusion and guiding surgical decision-making.

Case presentation: We report the case of a 29-year-old male presenting with a thunderclap headache, nausea, and vomiting, subsequently diagnosed with a fusiform aneurysm at the P2-P3 junction of the left posterior cerebral artery. The patient underwent a subtemporal approach with partial posterior petrosectomy for aneurysm clipping and remodeling. Initially, an STA-P3 and PITA-P3 bypass were considered; however, intraoperative ICG and FL-VAG confirmed sufficient retrograde collateral flow, allowing the bypass procedure to be avoided. Postoperative imaging demonstrated patent circulation in the occipitotemporal region without ischemic compromise.

Conclusion: This case highlights the crucial role of intraoperative fluorescence imaging in refining surgical strategies for complex aneurysm clipping. ICG and FL-VAG enhance surgical precision by providing real-time perfusion assessment, reducing the need for additional vascular interventions, and improving patient outcomes.

Objective: We examined the effectiveness of integrating three-dimensional (3D) visualization reconstruction technology with Problem-Based Learning (PBL) in the clinical teaching of resident physicians focusing on pheochromocytoma.

Methods: Fifty resident physicians specializing in urology at Peking Union Medical College Hospital were randomly divided into two groups over the period spanning January 2022 to January 2024: an experimental group and a control group. The experimental group underwent instruction utilizing a pedagogical approach that integrated 3D visualization reconstruction technology with PBL, while the control group used a traditional teaching model. A comparative analysis of examination performance and teaching satisfaction between both groups of resident physicians was conducted to assess the efficacy of the integrated 3D visualization and PBL teaching methods in clinical instruction.

Results: The experimental group demonstrated superior performance in both theoretical assessment and clinical skills evaluation, along with heightened levels of teaching satisfaction compared to the control group, with statistically significant differences (p < 0.05). Additionally, the experimental group exhibited markedly higher scores in both theoretical examinations and practical assessments compared to their counterparts in the control group (p < 0.05). The results of theoretical examinations for the experimental group and the control group were 92.15±3.22 and 81.09±4.46, respectively (< 0.0001). The results of practical examinations for the experimental group and the control group were 90.17±3.48 and 70.75±5.11, respectively (< 0.0001).

Conclusion: In the clinical teaching of training resident physicians specializing in urology for the management of pheochromocytoma, the integration of 3D visualization reconstruction technology with the PBL method significantly enhanced the teaching efficacy, improving both the quality of instruction and the level of satisfaction among participants.

Introduction: Breast cancer is a major cause of mortality among women globally. While mammography remains the gold standard for detection, its interpretation is often limited by radiologist variability and the challenge of differentiating benign and malignant lesions. The study explores the use of Auto- Sklearn, an automated machine learning (AutoML) framework, for breast tumor classification based on mammographic radiomic features.

Methods: 244 mammographic images were enhanced using Contrast Limited Adaptive Histogram Equalization (CLAHE) and segmented with Active Contour Method (ACM). Thirty-seven radiomic features, including first-order statistics, Gray-Level Co-occurance Matrix (GLCM) texture and shape features were extracted and standardized. Auto-Sklearn was employed to automate model selection, hyperparameter tuning and ensemble construction. The dataset was divided into 80% training and 20% testing set.

Results: The initial Auto-Sklearn model achieved an 88.71% accuracy on the training set and 55.10% on the testing sets. After the resampling strategy was applied, the accuracy for the training set and testing set increased to 95.26% and 76.16%, respectively. The Receiver Operating Curve and Area Under Curve (ROC-AUC) for the standard and resampling strategy of Auto-Sklearn were 0.660 and 0.840, outperforming conventional models, demonstrating its efficiency in automating radiomic classification tasks.

Discussion: The findings underscore Auto-Sklearn's ability to automate and enhance tumor classification performance using handcrafted radiomic features. Limitations include dataset size and absence of clinical metadata.

Conclusion: This study highlights the application of Auto-Sklearn as a scalable, automated and clinically relevant tool for breast cancer classification using mammographic radiomics.

Introduction: In nuclear medicine, Prostate-specific Membrane Antigen (PSMA) is a potential target for theranostics. Offering superior diagnostic accuracy to conventional imaging in prostate cancer (PCa), Gallium-68 labeled PSMA (⁶⁸Ga-PSMA) positron emission tomography/computed tomography (PET/CT) is considered the new standard of care in PCa management. Tumor cells identified as PSMA-avid on PET/CT imaging can be targeted and eliminated with PSMA-labeled Lutetium-177 (¹⁷⁷Lu-PSMA) therapy.

Case presentation: A sixty-eight years old patient who had metastatic castration-resistant PCa was reported in this study. Prior to receiving ¹⁷⁷Lu-PSMA therapy, the patient's PSA level was 358 ng/ml, and experienced extensive bone discomfort. Following ten cycles of ¹⁷⁷Lu-PSMA therapy, exceptional results were observed.

Conclusion: ¹⁷⁷Lu-PSMA therapy is likely to result in significantly better outcomes if first- or second-line treatments preserve the patient's bone marrow reserve or if the therapy is administered at earlier stages of the disease.