ABSTRACT
Wearable assistant devices play an important role in daily life for people with disabilities. Those who have hearing impairments may face dangers while walking or driving on the road. The major danger is their inability to hear warning sounds from cars or ambulances. Thus, the aim of this study is to develop a wearable assistant device with edge computing, allowing the hearing impaired to recognize the warning sounds from vehicles on the road. An EfficientNet-based, fuzzy rank-based ensemble model was proposed to classify seven audio sounds, and it was embedded in an Arduino Nano 33 BLE Sense development board. The audio files were obtained from the CREMA-D dataset and the Large-Scale Audio dataset of emergency vehicle sirens on the road, with a total number of 8756 files. The seven audio sounds included four vocalizations and three sirens. The audio signal was converted into a spectrogram by using the short-time Fourier transform for feature extraction. When one of the three sirens was detected, the wearable assistant device presented alarms by vibrating and displaying messages on the OLED panel. The performances of the EfficientNet-based, fuzzy rank-based ensemble model in offline computing achieved an accuracy of 97.1%, precision of 97.79%, sensitivity of 96.8%, and specificity of 97.04%. In edge computing, the results comprised an accuracy of 95.2%, precision of 93.2%, sensitivity of 95.3%, and specificity of 95.1%. Thus, the proposed wearable assistant device has the potential benefit of helping the hearing impaired to avoid traffic accidents.
Subject(s)
Hearing Loss , Wearable Electronic Devices , Humans , Ambulances , Hearing , Accidents, TrafficABSTRACT
STUDY DESIGN: A systematic review and meta-analysis. OBJECTIVE: To update the systematic review comparing the outcomes between surgical and non-surgical treatment for thoracolumbar burst fractures without neurological deficit. METHODS: We registered a protocol in PROSPERO (ID: CRD42021291769) and searched Medline, Embase, Web of Science, and Google Scholar databases. Surgical and non-surgical treatments were compared in patients with thoracolumbar burst fractures without neurological deficits. Predefined outcomes at ≥6 months included pain (defined as a visual analog scale [VAS] of 0-100), functional outcomes (Oswestry Disability Index [ODI] of 0-50 and Roland-Morris Disability Questionnaire [RMDQ] of 0-24), and kyphotic angulation. RESULTS: Nineteen studies involving 1056 patients were included in the analyses. For outcomes at ≥6 months, little to no difference was found in pain VAS score (mean difference, .95 [95% confidence interval {CI}, -6.02 to 7.92]; 827 participants; 15 studies; I2 = 92%), ODI (mean difference, -1.40 [95% CI, -5.11 to 2.31]; 446 participants; 7 studies; I2 = 79%), and RMDQ (mean difference, -.73 [95% CI, -5.13 to 3.66]; 216 participants; 5 studies; I2 = 77%). The kyphotic angulation in the surgery group was 6.35° lower than that in the non-surgery group (mean difference, -6.56° [95% CI, -10.26° to -2.87°]; 527 participants; ten studies; I2 = 86%). The trial sequential analysis indicated all outcomes reached adequate statistical power. The certainty of the evidence for all 4 outcomes was very low. For the analysis of minimally invasive procedures compared to traditional open surgeries, a statistically significant subgroup difference was found for VAS and ODI (P < .01 and P < .04, respectively). CONCLUSION: Surgical and non-surgical treatments showed little or no difference in outcomes at ≥6 months. This review provides a conclusion with adequate statistical power by including non-randomized studies. However, non-randomized studies also lowered the certainty of the evidence to a very low level.
ABSTRACT
PURPOSE: The incidence, pathogenesis, molecular pathways, and outcomes of colorectal cancer vary depending on the location of the tumor. This study aimed to compare the difference in tumor characteristics and the outcome between right-sided colon cancer and left-sided colorectal cancer (LCRC). MATERIALS AND METHODS: A total of 1503 patients with colorectal cancer who underwent surgery at the Taipei Veterans General Hospital between 2000 and 2010 were enrolled in this study. Right-sided colon cancer was defined as cancers in the cecum, ascending colon, and transverse colon, while LCRC was defined as cancers in the splenic flexure colon, descending colon, sigmoid colon, and rectum. The endpoint was overall survival. The mutations were detected via polymerase chain reaction and MASS array. The prognostic value was determined using the log-rank test and the Cox regression analysis. RESULTS: A total of 407 and 1096 cases were classified as right-sided colon cancer and LCRC, respectively. Compared to patients with LCRC, those with right-sided colon cancer had more mucinous type cancer (7.4% vs. 3.5%), poorly differentiated tumor (11.5% vs. 3.6%), and advanced tumor-node-metastasis stage. The risk for peritoneal tumor seeding was higher in the right-sided colon cancer group (12.8% vs. 5.7%). Overall survival was better in LCRC than in right-sided colon cancer ( P=0.036). CONCLUSIONS: In our study, right-sided colon cancer had a more advanced tumor stage, a higher risk of peritoneal metastasis, and a poorer outcome than LCRC. Moreover, right-sided colon cancer had more gene mutations in BRAF, KRAS, SMAD4, TGF-ß, PIK3CA, PTEN, AKT1, and high microsatellite instability.