Automatic Detection of Meniscus Tears Using Backbone Convolutional Neural Networks on Knee MRI.

BACKGROUND: Timely diagnosis of meniscus injuries is key for preventing knee joint dysfunction and improving patient outcomes because it decreases morbidity and facilitates treatment planning. PURPOSE: To train and evaluate a deep learning model for automated detection of meniscus tears on knee magnetic resonance imaging (MRI). STUDY TYPE: Bicentric retrospective study. SUBJECTS: In total, 584 knee MRI studies, divided among training (n = 234), testing (n = 200), and external validation (n = 150) data sets, were used in this study. The public data set MRNet was used as a second external validation data set to evaluate the performance of the model. SEQUENCE: A 3 T, coronal, and sagittal images from T1-weighted proton density (PD) fast spin-echo (FSE) with fat saturation and T2-weighted FSE with fat saturation sequences. ASSESSMENT: The detection system for meniscus tear was based on the improved YOLOv4 model with Darknet-53 as the backbone. The performance of the model was also compared with that of three radiologists of varying levels of experience. The determination of the presence of a meniscus tear from surgery reports was used as the ground truth for the images. STATISTICAL TESTS: Sensitivity, specificity, prevalence, positive predictive value, negative predictive value, accuracy, and receiver operating characteristic curve were used to evaluate the performance of the detection model. Two-way analysis of variance, Wilcoxon signed-rank test, and Tukey's multiple tests were used to evaluate differences in performance between the model and radiologists. RESULTS: The overall accuracies for detecting meniscus tears using our model on the internal testing, internal validation, and external validation data sets were 95.4%, 95.8%, and 78.8%, respectively. One radiologist had significantly lower performance than our model in detecting meniscal tears (accuracy: 0.9025 ± 0.093 vs. 0.9580 ± 0.025). DATA CONCLUSION: The proposed model had high sensitivity, specificity, and accuracy for detecting meniscus tears on knee MRIs. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.

End-to-end deep learning model for segmentation and severity staging of anterior cruciate ligament injuries from MRI.

PURPOSE: The purpose of this study was to develop a semi-supervised segmentation and classification deep learning model for the diagnosis of anterior cruciate ligament (ACL) tears on MRI based on a semi-supervised framework, double-linear layers U-Net (DCLU-Net). MATERIALS AND METHODS: A total of 297 participants who underwent of total of 303 MRI examination of the knee with fat-saturated proton density (PD) fast spin-echo (FSE) sequence in the sagittal plane were included. There were 214 men and 83 women, with a mean age of 37.46 ± 1.40 (standard deviation) years (range: 29-44 years). Of these, 107 participants had intact ACL (36%), 98 had partially torn ACL (33%), and 92 had fully ruptured ACL (31%). The DCLU-Net was combined with radiomic features for enhancing performances in the diagnosis of ACL tear. The different evaluation metrics for both classification (accuracy, sensitivity, accuracy) and segmentation (mean Dice similarity coefficient and root mean square error) were compared individually for each image class across the three phases of the model, with each value being compared to its respective value from the previous phase. Findings at arthroscopic knee surgery were used as the standard of reference. RESULTS: With the addition of radiomic features, the final model yielded accuracies of 90% (95% CI: 83-92), 82% (95% CI: 73-86), and 92% (95% CI: 87-94) for classifying ACL as intact, partially torn and fully ruptured, respectively. The DCLU-Net achieved mean Dice similarity coefficient and root mean square error of 0.78 (95% CI: 0.71-0.80) and 0.05 (95% CI: 0.06-0.07), respectively, when segmenting the three ACL conditions with pseudo data (P < 0.001). CONCLUSION: A dual-modules deep learning model with segmentation and classification capabilities was successfully developed. In addition, the use of semi-supervised techniques significantly reduced the amount of manual segmentation data without compromising performance.

Evaluation of Topical and Subconjunctival Injection of Hyaluronic Acid-Coated Nanoparticles for Drug Delivery to Posterior Eye.

Posterior eye diseases, such as age-related macular degeneration and diabetic retinopathy, are difficult to treat due to ineffective drug delivery to affected areas. Intravitreal injection is the primary method for posterior eye drug delivery; however, it is usually accompanied by complications. Therefore, an effective and non-invasive method is required. Self-assembling nanoparticles (NPs) made from gelatin-epigallocatechin gallate (EGCG) were synthesized (GE) and surface-decorated with hyaluronic acid (HA) for drug delivery to the retinal/choroidal area. Different HA concentrations were used to prepare NPs with negative (GEH-) or positive (GEH+) surface charges. The size/zeta potential and morphology of the NPs were characterized by a dynamic light scattering (DLS) system and transmission electron microscope (TEM). The size/zeta potential of GEH+ NPs was 253.4 nm and 9.2 mV. The GEH- NPs were 390.0 nm and -35.9 mV, respectively. The cytotoxicity was tested by adult human retinal pigment epithelial cells (ARPE-19), with the results revealing that variant NPs were non-toxicity at 0.2-50 µg/mL of EGCG, and that the highest amount of GEH+ NPs was accumulated in cells examined by flowcytometry. Topical delivery (eye drops) and subconjunctival injection (SCI) methods were used to evaluate the efficiency of NP delivery to the posterior eyes in a mouse model. Whole eyeball cryosections were used to trace the location of fluorescent NPs in the eyes. The area of fluorescent signal obtained in the posterior eyes treated with GEH+ NPs in both methods (eye drops: 6.89% and SCI: 14.55%) was the greatest when compared with other groups, especially higher than free dye solution (2.79%). In summary, GEH+ NPs can be transported to the retina by eye drops and SCI; in particular, eye drops are a noninvasive method. Furthermore, GEH+ NPs, characterized by a positive surface and HA decoration, could facilitate drug delivery to the posterior eye as a useful drug carrier.