Refined Detection and Classification of Knee Ligament Injury Based on ResNet Convolutional Neural Networks.

Currently, medical imaging has largely supplanted traditional methods in the realm of diagnosis and treatment planning. This shift is primarily attributable to the non-invasive nature, rapidity, and user-friendliness of medical-imaging techniques. The widespread adoption of medical imaging, however, has shifted the bottleneck to healthcare professionals who must analyze each case post-image acquisition. This process is characterized by its sluggishness and subjectivity, making it susceptible to errors. The anterior cruciate ligament (ACL), a frequently injured knee ligament, predominantly affects a youthful and sports-active demographic. ACL injuries often leave patients with substantial disabilities and alter knee mechanics. Since some of these cases necessitate surgery, it is crucial to accurately classify and detect ACL injury. This paper investigates the utilization of pre-trained convolutional neural networks featuring residual connections (ResNet) along with image-processing methods to identify ACL injury and differentiate between various tear levels. The ResNet employed in this study is not the standard ResNet but rather an adapted version capable of processing 3D volumes constructed from 2D image slices. Achieving a peak accuracy of 97.15% with a custom split, 96.32% through Monte-Carlo cross-validation, and 93.22% via five-fold cross-validation, our approach enhances the performance of three-class classifiers by over 7% in terms of raw accuracy. Moreover, we achieved an improvement of more than 1% across all types of evaluation. It is quite clear that the model's output can effectively serve as an initial diagnostic baseline for radiologists with minimal effort and nearly instantaneous results. This advancement underscores the paper's focus on harnessing deep learning for the nuanced detection and classification of ACL tears, demonstrating a significant leap toward automating and refining diagnostic accuracy in sports medicine and orthopedics.

Tumor Area Highlighting Using T2WI, ADC Map, and DWI Sequence Fusion on bpMRI Images for Better Prostate Cancer Diagnosis.

Prostate cancer is the second most common cancer in men worldwide. The results obtained in magnetic resonance imaging examinations are used to decide the indication, type, and location of a prostate biopsy and contribute information about the characterization or aggressiveness of detected cancers, including tumor progression over time. This study proposes a method to highlight prostate lesions with a high and very high risk of being malignant by overlaying a T2-weighted image, apparent diffusion coefficient map, and diffusion-weighted image sequences using 204 pairs of slices from 80 examined patients. It was reviewed by two radiologists who segmented suspicious lesions and labeled them according to the prostate imaging-reporting and data system (PI-RADS) score. Both radiologists found the algorithm to be useful as a "first opinion", and they gave an average score on the quality of the highlight of 9.2 and 9.3, with an agreement of 0.96.

Is the Ellipsoid Formula Reliable in Prostate MRI?

Our study seeks to study the accuracy of the ellipsoidal formula in prostate MRI of different sizes and to establish the limits of its use. The study included 31 patients with a well-visualized, intact prostatic capsule, excluding malignantly transformed prostates, as well as treated prostates, in which the contrast between the prostatic capsule and parenchyma is reduced. Each patient's prostatic volume was recalculated according to the ellipsoidal formula, and then it was compared with the prostatic volume calculated by the segmentation method. The two calculated volumes were similar, in some cases almost identical, with a slight tendency to underestimate prostate volume below 100cm3, in total in 18 cases, on average by 7.6% (+/-6%), overestimation of those with a volume over 100cm3, a total of 13 cases, on average by 3.2% (+/-2.5%), and of all, in 4 cases the difference between the two formulas was below 1%. There was no statistical difference between the two variables, Student's t-test p-value=0.039. With a precision of 92% (+/-6%), the ellipsoidal formula can be considered accurate when it is correctly performed, but if we take into account the importance that PSA density is starting to have in diagnosis, treatment and follow-up, the calculation of a secondary value through the segmentation method or high-precision software can be motivated when the ellipsoidal formula returns a value close to a threshold.