OBJECTIVE To construct a remotely contactless intelligent sleep monitoring system(RCISMS),which could be used to assist the titration of Mandibular advancement device(MAD)for the patients of obstructive sleep apnea-hypopnea syndrome(OSAHS),and to verify the effectiveness of the system.METHODS RCISMS was constructed by multidisciplinary medical-engineering cooperation of clinical medicine,biosensing technology,and artificial intelligence;And the diagnostic and therapeutic efficacy evaluation of RCISMS was improved through clinical comparative tests.From December 2023 to May 2024,A total of 176 patients who would receive polysomnography(PSG)evaluation were prospectively collected in the Department of Otolaryngology,Beijing Jishuitan Hospital,Capital Medical University.All the patients met the inclusion criteria and underwent synchronized sleep monitoring using RCISMS and PSG.Through repeatedly comparison with the'gold standard'PSG,the intelligent algorithm of Apnea-hypopnea index(AHI)which is the diagnosis and efficacy of OSAHS was improved in RCISMS.The consistency and correlation of the results between RCISMS and PSG were statistically analyzed.RESULTS RCISMS can continuously monitor the patient's natural sleep in a remote,non-direct contact way,and can upload data in real time,analyze and diagnose intelligently,and the results can be distributed to the doctor and the patient simultaneously.Among the parameters monitored by RCISMS,the AHI,lowest peripheral oxygen saturation(LSpO2),mean peripheral oxygen saturation(MSpO?),sleep efficiency,and the proportion of light sleep stage showed strong correlation with the corresponding PSG parameters(all P<0.05).The LSpO2 and the proportion of light sleep stage showed no statistically significant differences between the two monitoring methods.The consistency test between RCISMS and PSG for AHI classification yielded a Kappa value of 0.627(P<0.001),indicating moderate agreement.The efficacy evaluation of RCISMS for MAD treatment assessment demonstrated 73.8%sensitivity and 100%specificity for OSAHS detection,which helped to precisely identify patients who had achieved full recovery at the endpoint of MAD titration.RCISMS demonstrated sensitivities of 73.8%,84.8%,and 92.7%for detecting mild,moderate,and severe OSAHS respectively,with corresponding specificities of 78.4%,91.6%,and 95.9%.This system can objectively reflect therapeutic changes during MAD titration.CONCLUSION This study established the RCISMS,which provided a novel titration procedure for the MAD titration of OSAHS patients,and might potentially overcome existing limitations in current clinical MAD titration procedures.

Objective To explore the value of dual-source dual-energy CT (DSDE-CT) in differentiating extravasation of iodine contrast agents from secondary hemorrhage after revascularization in acute ischemic stroke. Methods 46 acute ischemic stroke patients following intra-arterial thrombolysis were examined with DSDE-CT within 2 hours after the procedure. Simultaneous imaging at 80 kV/392 mA and 140 kV/196 mA was employed, and then mixed images, virtual unenhanced non-contrast images and iodine overlay maps were calculated. Mixed images alone, as conventional CT, and DUDE-CT interpretations were assessed separately by two radiologists and compared with follow-up CT. Results 6 of 34 patients were negative cases proven by CT without high density, and another 28 cases were proven positive cases with 3 cases of cerebral hemorrhage, 21 cases of contrast agent extravasation, and the remaining 4 cases of combined cerebral hemorrhage and contrast agent extravasation. The sensibility, specificity, positive predictive value, negative predictive value and accuracy of mixed imaging alone in diagnosing hemorrhage was 66.67%, 100%, 1005, 96.15% and 96.43%, while the sensibility, specificity, positive predictive value, negative predictive value and accuracy of hemorrhage with DUDE-CT was 100%, 96%, 75%, 100% and 96.43% . The diagnostic accuracy of superimposed fusion images for intracranial hemorrhage, extravasation of contrast agent and hemorrhage with extravasation of contrast agent was relatively high, and the difference was statistically significant (P < 0.05). The diagnostic accuracy of superimposed fusion images consistent with clinical follow-up was significantly higher (Kappa=0.815),as compared with that of mixed imaging alone (Kappa=0.0.564). Conclusion DUDE-CT has great value in differentiating hemorrhage from iodinated contrast after intra-arterial thrombolysis in acute ischemic stroke.