Current Medical Imaging Reviews最新文献

Background: Spinal image denoising plays a vital role in the accurate diagnosis of disc herniation (DH).

Objective: Traditional denoising algorithms perform less due Limited Directional Selectivity problem and do not adequately capture directional information in pixels. Traditional algorithms' edge representation and texture details are insufficient for the earlier detection of DH. Limited Directional Selectivity leads to inaccurate diagnosis and classification of Disc Herniation (DH) stages. The DH stages are (i) Degeneration (ii) Prolapse (iii) Extrusion and (iv) Sequestration. Moreover, detection of DH size below 2mm using MR image is the major problem.

Methods: To solve the above problem, spinal cord MR images fed to the proposed Parrot optimization tuned Denoising Convolutional Neural Network (Po- DnCNN) algorithm for perspective enhancement of nucleus pulposus region in the spinal cord, vertebrae. The perspective enhancement of Spinal cord image led to the accurate classification of stages and earlier detection of DH by using the proposed Hippopotamus optimization- Fast Hybrid Vision Transformer (Ho-FastViT) algorithm. For this study, spinal cord MR images are obtained from the Grand Challenge website - SPIDER dataset.

Results: The proposed Po-DnCNN method and Ho-FastViT results are analysed quantitatively and qualitatively based on the edge, contrast, classification of the stage, and enhancement of the projected nucleus pulposus region in the spinal cord and vertebrae. The predicted DH results using the proposed method are compared with the manual Pfirrman Grade value of the spinal card method.

Conclusion: Proposed method is better than traditional methods for earlier detection of DH. Po-DnCNN and Ho-FastViat methods give high accuracy of about 98% and 97% compared to traditional methods.

Background: Early diagnosis of prostate cancer can improve the survival rate of patients on the premise of high-quality images. The prerequisite for early diagnosis is high-quality images. ZOOMit is a method for high-resolution, zoomed FOV imaging, allowing diffusion-weighted images with high contrast and resolution in short acquisition times. RESOLVE DWI is an advanced MRI technique developed to obtain high-resolution diffusionweighted images with reduced susceptibility-related artifacts.

Objective: This study aimed to compare the image quality of conventional single-shot Echo-planar Imaging (ss-EPI), Diffusion-weighted Imaging (DWI), zoomed FOV imaging (ZOOMit) DWI, and readout segmentation of long variable echo-trains (RESOLVE) DWI sequences for prostate imaging, and optimize the strategy to obtain high-quality Magnetic Resonance Imaging (MRI) in order to discriminate malignant and benign prostate diseases.

Methods: Fifty-one patients were enrolled, including 31 with prostate cancer, 11 with prostate benign disease, and 9 with bladder cancer. Patients underwent MRI scans using T2-weighted (T2W), ss-EPI DWI, ZOOMit DWI, and RESOLVE DWI (b = 0, 50, 1400 s/mm2) sequences using a 3.0T MRI scanner. Subjective scores of image quality were evaluated by two independent radiologists. Differences in the subjective scores and objective parameters among the three sequences were compared. The agreement and consistency between the findings of the two raters were evaluated with Kappa or Intra-class Correlation Coefficient (ICC). Receiver Operating Characteristic (ROC) curves were used to distinguish malignant and benign prostate disease.

Results: The agreement of subjective scores of 51 patients was high or moderate between the two radiologists (kappa: 0.529-0.880). ZOOMit displayed the highest clarity and the lowest distortion and artifacts compared to ss-EPI and RESOLVE. The two radiologic technicians obtained moderate or high consistency of objective measurement (ICC: 0.527-0.924). In the ROC analysis, ADCmean and Prostate Imaging Reporting and Data System (PI-RADS) scores for three sequences were comparable in differentiating prostate cancer from benign prostate disease (all p>0.05), in which ZOOMit indicated the highest Area Under the Curve (AUC) (0.930 and 0.790, respectively).

Conclusion: Compared to the other two sequences, ZOOMit can be deemed preferable to improve prostate MRI diffusion imaging as it has exhibited the highest AUC in identifying prostate cancer.

Background: Laparoscopically Harvested Pedicled Omental Flap [LHPOF] has become a viable option for breast reconstruction due to advancements in minimally invasive techniques, offering benefits like reduced postoperative pain and minimal scarring.

Case presentation: This study examines the imaging findings in eight patients who underwent breast reconstruction using a LHPOF. Imaging modalities, including mammography, ultrasonography, MRI, and CT, consistently showed reconstructed breasts with fat replacing glandular tissue and numerous internal vessels. One case of recurrence was detected, demonstrating the efficacy of conventional surveillance imaging studies in facilitating the detection of recurrences.

Conclusion: This is the first report detailing imaging findings of breast reconstruction using an LHPOF, including a recurrence case. Understanding these imaging results is crucial for effective surveillance in breast cancer patients with omental flap reconstruction.

Purpose: This study aims to evaluate the clinical efficacy of spectral dual-energy detector computed tomography (SDCT) and its associated parameters in diagnosing acute pulmonary embolism (APE).

Methods: Retrospective analysis of imaging data from 86 APE-diagnosed patients using SDCT was conducted. Virtual monoenergetic images (VMIs) at 40, 70, and 100 KeV, Iodine concentration (IC) maps, Electron Cloud Density Map (ECDM), Effective atomic number (Z-eff) maps, and Hounsfield unit attenuation plots (VMI slope) were reconstructed from pulmonary artery phase CT images. The subtraction (SUB) and ratios of VMIs were calculated, and two experienced radiologists evaluated the patients. The Mann-Whitney U test assessed the parameter ability to differentiate between normal and obstructed lung fields and detect emboli in the pulmonary artery. Receiver Operating Characteristic Curves (ROC) were generated for performance evaluation.

Results: Significant differences (p<0.001) in 40KeV, Ratio, SUB, and Z-eff were found between normal and embolized lung fields. Logistic regression demonstrated good detection performance for Z-eff (AUC=0.986), SUB (AUC=0.975), and IC (AUC=0.974). Parameters such as 40KeV (AUC=0.990), 70KeV (AUC=0.980), 100KeV (AUC=0.962), SUB (AUC=0.990), Z-eff (AUC=0.999), and IC (AUC=1.000) exhibited good detection capabilities for identifying emboli in the pulmonary artery.

Conclusion: SDCT facilitates the identification of diseased lung fields and localization of emboli in the pulmonary artery, thereby improving diagnostic efficiency in APE.

Background: Extrahepatic Common Bile Duct Obstruction (EHBDO) is a serious condition that requires accurate diagnosis for effective treatment. Magnetic Resonance Cholangiopancreatography (MRCP) is a widely used noninvasive imaging technique for visualizing bile ducts, but its interpretation can be complex.

Objective: This study aimed to develop a deep learning-based classification model that integrates MRCP images and clinical parameters to assist radiologists in diagnosing EHBDO more accurately.

Methods: A total of 465 patients with clinical data were included, of whom 143 also had MRCP images. Missing clinical values were addressed through data imputation. Object detection techniques were used to isolate the common bile duct region in the MRCP images. A multimodal deep learning fusion model was developed by combining the extracted imaging features with selected clinical parameters. To account for the varying significance of different features, a weighted loss function was applied. The performance of the fusion model was compared to that of single-modality approaches (using only MRCP images or clinical data), specifically the accuracy, sensitivity, specificity, and Area Under The Curve (AUC).

Results: The performance of the proposed deep learning fusion model was superior to that of models using only MRCP images or clinical parameters. The fusion model achieved an accuracy of 89.8%, AUC of 90.4%, sensitivity of 81.8%, and specificity of 95.7% in diagnosing EHBDO. By integrating MRCP imaging data and clinical parameters, the proposed deep learning model significantly enhanced the accuracy of EHBDO diagnosis.

Conclusion: This proposed multimodal approach outperformed traditional single-modality methods, presenting a valuable tool for improving the diagnostic accuracy of bile duct obstruction.

Background: Axitinib is a selective inhibitor of vascular endothelial growth factor receptors and is used in the treatment of many malignancies. Herein, we reported a rare case with axitinib-induced retinal changesand associated toxicity.

Case presentation: A forty-five-year-old female presented with blurred vision who had been taking 7 mg of Axitinib bid for 5 months. Initial Best Corrected Visual Acuity (BCVA) was 20/32 at the right and counting fingers at the left eye. Funduscopic examination revealed bilaterally widespread intraretinal hemorrhages, cotton-wool spots, and hard exudates with a star-like appearance at the macula. The optical coherence tomography revealed central macular edema. There was hyperreflective edema in the inner layers, exudates in the middle retinal layers, and subfoveal subretinal fluid. Fundus fluorescein angiography revealed localized ischaemic findings in the early phase and multifocal perivascular ink-blot fluorescein leakage in the middle and late phases. Axitinib treatment was discontinued immediately, and at the third month of follow-up, the macular edema and fundus findings improved with a final BCVA of 20/20 at the right and 20/32 at the left eye.

Conclusion: Considering the ocular side effects of the patients receiving axitinib is crucial to prevent any potentially persistent visual loss.

Objective The purpose of this study was to establish the efficacy of ultrasound (US)-guided hydrodilatation of the glenohumeral joint, in conjunction with corticosteroid injection, in alleviating pain and improving shoulder joint adhesion among patients with primary frozen shoulder (FS). Background: FS, also known as adhesive capsulitis, is a pathological condition characterized by pain and potential functional impairment. The natural progression of FS involves three distinct stages: freezing, frozen, and thawing. Chronic pain in FS patients can lead to aseptic inflammation, thickening of fibroblasts, and an abundance of type I and III collagen fibers in the vicinity of the glenohumeral joint, ligaments, and tendons. This condition significantly impacts patients' quality of life. Methods A total of 200 FS patients were enrolled in this study. All participants underwent US-guided hydrodilatation of the glenohumeral joint, combined with corticosteroid injection, at our department. Pre- and post-treatment (1 year) ultrasound measurements were recorded for the thickness of the axillary recess capsule (ARC), coracohumeral ligament (CHL), and subacromial bursa. Additionally, the numerical rating scale (NRS) and Constant-Murley score (CMS) were assessed to evaluate pain intensity and shoulder function, respectively. Results Prior to the commencement of treatment, the measurements indicated a thickness of 4.8±2.3 mm for the ARC, 4.2±1.7 mm for the CHL, and 3.9±1.4 mm for the subacromial bursa. Additionally, the preoperative assessment using the NRS scale for pain yielded a score of 6.4±2.0, while the CMS score for the joint function was 35.8±8.5. Following one year of treatment, a notable decrease was observed in the thickness of ARC, CHL, and subacromial bursa. Furthermore, significant improvements were recorded in both the pain NRS score and the CMS score. Conclusion US-guided hydrodilatation of the glenohumeral joint, in combination with corticosteroid injection, may help improve the symptom and function of FS.

Objective: The primary objective of this comparative investigation was to examine the qualitative attributes of image reconstructions utilizing two distinct algorithms, namely OSEM and HYPER Iterative, in total-body 18F- FDG PET/CT under various acquisition durations and injection activities.

Methods: An initial assessment was executed using a NEMA phantom to compare image quality engendered by OSEM and HYPER Iterative algorithms. Parameters such as BV, COV, and CRC were meticulously evaluated. Subsequently, a prospective cohort study was conducted on 50 patients, employing both reconstruction algorithms. The study was compartmentalized into distinct acquisition time and dosage groups. Lesions were further categorized into three size-based groups. Quantifiable metrics including SD of noise, SUV_max, SNR, and TBR were computed. Additionally, the differences in values, namely ΔSUV_max, ΔTBR, %ΔSUV_max, %ΔSD, and %ΔSNR, between OSEM and HYPER Iterative algorithms were also calculated.

Results: The HYPER Iterative algorithm showed reduced BV and COV compared to OSEM in the phantom study, with constant acquisition time. In the clinical study, lesion SUV_max, TBR, and SNR were significantly elevated in images reconstructed using the HYPER Iterative algorithm in comparison to those generated by OSEM (p < 0.001). Furthermore, an amplified increase in SUV_max was predominantly discernible in lesions with dimensions less than 10 mm. Metrics such as %ΔSNR and %ΔSD in HYPER Iterative exhibited improvements correlating with reduced acquisition times and dosages, wherein a more pronounced degree of enhancement was observable in both ΔSUV_max and ΔTBR.

Conclusion: The HYPER Iterative algorithm significantly improves SUV_max and reduces noise level, with particular efficacy in lesions measuring ≤ 10 mm and under conditions of abbreviated acquisition times and lower dosages.

Background: In the realm of cancer studies,the differences among the biological behavior of oligometastatic prostate cancer (OPCa), localized prostate cancer (LPCa), and widely prostate cancer (WPCa) are still unclear.

Objectives: The purpose of our study was to assess the clinical and intravoxel incoherent motion (IVIM) parameters of tumor burden in OPCa. In addition, the correlation between clinical and IVIM parameters and the prostate-specific antigen nadir (PSAN) and time to nadir (TTN) during initial androgen deprivation therapy (ADT) in OPCa was explored. It was found that the IVIM parameters could effectively differentiate LPCa and WPCa, as well as LPCa and OPC. Moreover, Gleason score (GS) was positively correlated with PSAN, while prostate volume was positively correlated with TTN.

Methods: About 54 patients were included in this retrospective study (mean age=74±7.4 years). ADC, D, D*, and f were acquired according to the biexponential Diffusion Weighted Imaging (DWI) model. The Kruskal-Wallis test was used to test the differences in clinical and IVIM parameters among the three groups. The Receiver Operating Characteristic (ROC) curve was used to evaluate the discrimination abilities. The Area Under the Curve (AUC) was compared using the DeLong test. Furthermore, Spearman correlation analysis was performed to assess the correlation between clinical and IVIM parameters of PSAN and TTN during initial ADT with OPCa.

Results: There were significant differences among the three groups observed for age, PSA, GS, ADC, D and D* values (P<0.05). Multi-parameter pairwise comparison results showed that significant differences between LPCa and WPCa were observed for the age, PSA, GS, ADC, D and D* values (P<0.05). However, D* was different between the LPCa and OPCa groups (P=0.032). GS showed a significant positive correlation with PSAN (Rho=0.594, P=0.042), and prostate volume showed a significant positive correlation with TTN (Rho=0.777, P=0.003).

Conclusions: The IVIM parameters can effectively differentiate LPCa and WPCa, as well as LPCa and OPCa. Moreover, there was a certain trend in their distribution, which could reflect the tumor burden of PCa.

Objective: To develop and validate a radiomics-clinical nomogram for the detection of the acquired T790M mutation in patients with advanced non-small cell lung cancer (NSCLC) with resistance after the duration of first-line epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitor (TKI) treatment.

Materials and methods: Thoracic CT was collected from 120 advanced NSCLC patients who suffered progression on first- or second-generation TKIs. Radiomics signatures were retrieved from the entire tumor. Pearson correlation and the least absolute shrinkage and selection operator (LASSO) regression method were adopted to choose the most suitable radiomics features. Clinical and radiological factors were assessed using univariate and multivariate analysis. Three Machine Learning (ML) models were constructed according to three classifiers, including Logistic Regression (LR), Support Vector Machine (SVM), and RandomForest (RF), combining clinical and radiomic features. A nomogram combining clinical features and the rad score signature was built. The predictive ability of the nomogram was assessed by the ROC curve, calibration curve, and decision curve analysis (DCA).

Results: Multivariate regression analysis showed that two clinicopathological characteristics and two radiological features were highly correlated with the acquired T790M mutation, including the progression-free survival (PFS) of first-line EGFR TKIs (P = 0.029), the initial EGFR profile (P = 0.01), vascular convergence (P = 0.043), and air bronchogram (P = 0.030). The AUCs of clinical, radiomics, and combined models using RF classifiers for T790M mutation detection were 0.951 (95% confidence interval [CI] 0.911,0.991), 0.917 (95%CI 0.856,0.978), and 0.961 (95%CI 0.927,0.995) in the training cohort, respectively, higher than those of other classifier models.The calibration curve and Hosmer-Lemeshow Test showed good calibration power, and the DCA demonstrated a significant net benefit.

Conclusion: A radiomics-clinical nomogram based on CT radiomics proved valuable in non-invasively and efficiently predicting the acquired T790M mutation in patients who suffered progression on first-line TKIs.