ABSTRACT
BACKGROUND: Traumatic knee injuries are challenging to diagnose accurately through radiography and to a lesser extent, through CT, with fractures sometimes overlooked. Ancillary signs like joint effusion or lipo-hemarthrosis are indicative of fractures, suggesting the need for further imaging. Artificial Intelligence (AI) can automate image analysis, improving diagnostic accuracy and help prioritizing clinically important X-ray or CT studies. OBJECTIVE: To develop and evaluate an AI algorithm for detecting effusion of any kind in knee X-rays and selected CT images and distinguishing between simple effusion and lipo-hemarthrosis indicative of intra-articular fractures. METHODS: This retrospective study analyzed post traumatic knee imaging from January 2016 to February 2023, categorizing images into lipo-hemarthrosis, simple effusion, or normal. It utilized the FishNet-150 algorithm for image classification, with class activation maps highlighting decision-influential regions. The AI's diagnostic accuracy was validated against a gold standard, based on the evaluations made by a radiologist with at least four years of experience. RESULTS: Analysis included CT images from 515 patients and X-rays from 637 post traumatic patients, identifying lipo-hemarthrosis, simple effusion, and normal findings. The AI showed an AUC of 0.81 for detecting any effusion, 0.78 for simple effusion, and 0.83 for lipo-hemarthrosis in X-rays; and 0.89, 0.89, and 0.91, respectively, in CTs. CONCLUSION: The AI algorithm effectively detects knee effusion and differentiates between simple effusion and lipo-hemarthrosis in post-traumatic patients for both X-rays and selected CT images further studies are needed to validate these results.
Subject(s)
Artificial Intelligence , Hemarthrosis , Knee Injuries , Tomography, X-Ray Computed , Humans , Knee Injuries/diagnostic imaging , Knee Injuries/complications , Tomography, X-Ray Computed/methods , Female , Male , Retrospective Studies , Hemarthrosis/diagnostic imaging , Hemarthrosis/etiology , Middle Aged , Adult , Algorithms , Aged , Exudates and Transudates/diagnostic imaging , Aged, 80 and over , Young Adult , Adolescent , Radiographic Image Interpretation, Computer-Assisted/methods , Knee Joint/diagnostic imaging , Sensitivity and SpecificityABSTRACT
PURPOSE: Apparent diffusion coefficient (ADC) values in the developing fetus provide valuable information on the diagnosis and prognosis of prenatal brain pathologies. Normative ADC data has been previously established in 1.5 T MR scanners but lacking in 3.0 T scanners. Our objective was to measure ADC values in various brain areas in a cohort of normal singleton fetuses scanned in a 3.0 T MR scanner. METHODS: DWI (diffusion-weighted imaging) was performed in 47 singleton fetuses with normal or questionably abnormal results on sonography followed by normal structural MR imaging. ADC values were measured in cerebral lobes (frontal, parietal, temporal lobes), basal ganglia, and pons. Regression analysis was used to examine gestational age-related changes in regional ADC. RESULTS: Median gestational age was 30.1 weeks (range, 26-34 weeks). There was a significant effect of region on ADC values, whereby ADC values were highest in cerebral lobes (parietal > frontal > temporal lobes), compared with basal ganglia. The lowest values were found in the pons. On regression analysis, there was a decrease in ADC values in basal ganglia and pons with increasing gestational age. ADC values in frontal, parietal, and temporal lobes were stable in our cohort. CONCLUSION: Regional brain ADC values in 3.0 T scanners are comparable with previously reported values in 1.5 T scanners, with similar changes over gestational age. Using 3.0 T scanners is increasing worldwide. For fetal imaging, establishing normal ADC values is critical as DWI enables a sensitive and quantitative technique to evaluate normal and abnormal brain development.
