RESUMO
Sleep disorders, prevalent in the general population, present significant health challenges. The current diagnostic approach, based on a manual analysis of overnight polysomnograms (PSGs), is costly and time-consuming. Artificial intelligence has emerged as a promising tool in this context, offering a more accessible and personalized approach to diagnosis, particularly beneficial for under-served populations. This is a systematic review of AI-based models for sleep disorder diagnostics that were trained, validated, and tested on diverse clinical datasets. An extensive search of PubMed and IEEE databases yielded 2114 articles, but only 18 met our stringent selection criteria, underscoring the scarcity of thoroughly validated AI models in sleep medicine. The findings emphasize the necessity of a rigorous validation of AI models on multimodal clinical data, a step crucial for their integration into clinical practice. This would be in line with the American Academy of Sleep Medicine's support of AI research.
RESUMO
PURPOSE: Mastoid air cell abnormalities in the form of hyperintense T2 fluid signal have been reported in cases of acute Cerebral Venous Thrombosis (CVT) without otologic infection and have been hypothesized to be a result of venous congestion rather than infectious mastoiditis. The aim of this study was to investigate a link between the spectrum of mastoid abnormalities and clot burden in patients with acute CVT. METHODS: A retrospective study of adult patients admitted to the National Institute of Mental Health and Neurosciences between 2016 and 2023 who were diagnosed with acute CVT and had no clinical evidence of active or recent ear infections was conducted. Pre- and post-contrast MR Images were analyzed to identify the dural sinuses and/or cerebral veins involved and the presence of fluid signal in the mastoid. Fluid signal in the mastoid was graded from 0 to 3 as described by Shah et al- no fluid signal (grade 0), thin curvilinear hyperintensities (grade 1), thick crescenteric hyperintensities (grade 2), and complete hyperintensity (grade 3). Clot Burden Score (CBS) was calculated by assigning one point for each sinus involved, one point for extension of thrombus into the intracranial Internal Jugular Vein (IJV), one point for thrombosis of cortical veins and one point for thrombosis of deep cerebral veins. RESULTS: A total of 89 patients with acute CVT were included in the final analysis. Median time from presentation to MRI was 2 days (range 0-13). 51 patients (57.3%) had fluid signal in the mastoid air cells on T2-weighted images, of whom 33 showed mucosal contrast enhancement. Higher grade of fluid signal in the mastoid was present ipsilateral to the side of venous thrombosis in 59 out of 60 patients with posterior fossa CVT. CBS was significantly different between patients with different grades of fluid signal (p = 0.002). Grade 2-3 fluid signal was associated with higher clot burden (CBS > 3) in both the entire study population (n = 89) - OR = 8.281, 95 %CI: 2.758-24.866 (p < 0.001) and among patients with posterior fossa CVT - OR = 4.375, 95 %CI: 1.320-14.504 (p = 0.016). Among patients with posterior fossa CVT, grade 2-3 fluid signal was associated with left sided transverse and/or sigmoid sinus thrombosis - OR = 5.600, 95 %CI: 1.413-22.188 (p = 0.014), and extension of thrombosis into the IJV - OR = 4.606, 95 %CI: 1.162-18.262 (p = 0.030). CONCLUSION: T2 fluid signal in the mastoid is associated with venous congestion in adults with acute CVT without evidence of otologic infection. Moderate-to-severe T2 fluid signal in the mastoid air cells is associated with increased clot burden.
