Deep learning radiomics based on ultrasound images for the assisted diagnosis of chronic kidney disease.

AIM: This study aimed to explore the value of ultrasound (US) images in chronic kidney disease (CKD) screening by constructing a CKD screening model based on grey-scale US images. METHODS: According to the CKD diagnostic criteria, 1049 patients from Tongde Hospital of Zhejiang Province were retrospectively enrolled in the study. A total of 4365 renal US images were collected from these patients. Convolutional neural networks were used for feature extractions and a screening model was constructed by fusing ResNet34 and texture features to identify CKD and its stage. A comparative analysis was performed to compare the diagnosis results of the model with physicians. RESULTS: When diagnosing CKD or non-CKD, the receiver operating characteristic curve (AUC) of our model was 0.918 and that of the senior physician group was 0.869 (p < .05). For the diagnosis of CKD stage, the AUC of our model for CKD G1-G3 was 0.781, 0.880, and 0.905, respectively, while the AUC of the senior physician group for CKD G1-G3 was 0.506, 0.586, and 0.796, respectively; all differences were statistically significant (p < .05). The diagnostic efficiency of our model for CKD G4 and G5 reached the level of the senior physicians group. Specifically, the AUC of our model for CKD G4-G5 was 0.867 and 0.931, respectively, while the AUC of the senior physician group for CKD G4-G5 was 0.838 and 0.963, respectively (all p > .05). CONCLUSIONS: Our deep learning radiomics model is more effective than senior physicians in the diagnosis of early CKD.

Associations of sleep duration and quality with incident cardiovascular disease, cancer, and mortality: a prospective cohort study of 407,500 UK biobank participants.

OBJECTIVE: Few studies have investigated the associations of sleep duration and sleep quality with incident cardiovascular diseases (CVDs), cancer, and mortality in the same large population. This study aimed at estimating the independent risk factors of long or short sleep durations and several typical characteristics of poor sleep quality for incident CVDs, cancer, and mortality. METHODS: In this prospective cohort study, 407 500 individuals were enrolled. Cox proportional hazards models were used to calculate the adjusted hazard ratios and 95% confidence intervals (HR, 95%CI) of associations of sleep duration and quality with incident CVDs, cancer, and mortality. RESULTS: Compared with the sleep duration of 7 h, sleep duration of ≤5 h and ≥9 h were both associated with higher risk of all-cause mortality (HR = 1.25, 95% CI: 1.16-1.34 and HR = 1.30, 95% CI: 1.22-1.38, respectively), CVD mortality (HR = 1.27, 95% CI: 1.09-1.49 and HR = 1.32, 95% CI: 1.16-1.50, respectively), and CVD incidence (HR = 1.23, 95% CI: 1.16-1.31 and HR = 1.08, 95% CI: 1.02-1.15, respectively). Additionally, long sleep duration (≥9 h) was associated with a higher risk of cancer mortality (HR = 1.19, 95% CI: 1.10-1.30) and cancer incidence (HR = 1.08, 95% CI: 1.04-1.12). Moreover, CVD incidence was significantly associated with snoring, insomnia and narcolepsy, increasing the risk by 7%, 26%, and 20%, respectively. CONCLUSION: Long sleep durations may substantially increase the risk of mortality and morbidity. Snoring, insomnia, and narcolepsy were independent risk factors for incident CVD.

A Mobile-Based Intervention for Glycemic Control in Patients With Type 2 Diabetes: Retrospective, Propensity Score-Matched Cohort Study.

BACKGROUND: Mobile-based interventions appear to be promising in ameliorating huge burdens experienced by patients with type 2 diabetes. However, it is unclear how effective mobile-based interventions are in glycemic management of patients with type 2 diabetes based on real-world evidence. OBJECTIVE: This study aimed to evaluate the effectiveness of a mobile-based intervention on glycemic control in patients with type 2 diabetes based on real-world population data. METHODS: This retrospective, propensity score-matched cohort study analyzed longitudinal data from a clinical electronic health database. The study population included 37,913 patients with type 2 diabetes at cohort entry between October 1, 2016, and July 31, 2018. A total of 2400 patients were matched 1:1, using propensity score matching, into the usual care and mobile health (mHealth) groups. The primary outcomes of glycemic control included control rates of glycated hemoglobin (HbA1c), fasting blood glucose (FBG), and postprandial 2-hour blood glucose (P2BG). Mean values and variation trends of difference with 95% CI were the secondary outcomes. The general linear model was used to calculate repeated-measures analyses of variance to examine the differences between the two groups. Subgroup and sensitivity analyses were performed. RESULTS: Of the 2400 patients included in the analysis, 1440 (60.00%) were male and the mean age was 52.24 years (SD 11.56). At baseline, the control rates of HbA1c, FBG, and P2BG in the mHealth and usual care groups were 45.75% versus 47.00% (P=.57), 38.03% versus 32.76% (P=.07), and 47.32% versus 47.89% (P=.83), respectively. At the 3-, 6-, 9-, and 12-month follow-ups, the mHealth group reported higher control rates of HbA1c than did the usual care group: 69.97% versus 46.06% (P<.001), 71.89% versus 61.24% (P=.004), 75.38% versus 53.44% (P<.001), and 72.31% versus 46.70% (P<.001), respectively. At the four follow-up sessions, the control rates of FBG in the mHealth and usual care groups were statistically different: 59.24% versus 34.21% (P<.001), 56.61% versus 35.14% (P<.001), 59.54% versus 34.99% (P<.001), and 59.77% versus 32.83% (P<.001), respectively. At the four follow-up sessions, the control rates of P2BG in the mHealth group were statistically higher than in the usual care group: 79.72% versus 48.75% (P<.001), 80.20% versus 57.45% (P<.001), 81.97% versus 54.07% (P<.001), and 76.19% versus 54.21% (P=.001), respectively. At the four follow-up sessions, the percentages of HbA1c reduction in the mHealth group were 8.66% (95% CI 6.69-10.63), 10.60% (95% CI 8.66-12.54), 10.64% (95% CI 8.70-12.58), and 8.11% (95% CI 6.08-10.14), respectively. At the four follow-up sessions, the percentages of P2BG reduction in the mHealth group were 8.44% (95% CI 7.41-10.73), 17.77% (95% CI 14.98-20.23), 16.23% (95% CI 13.05-19.35), and 16.91% (95% CI 13.17-19.84), respectively. Starting from the sixth month, the mean HbA1c and P2BG values in the two groups increased slightly. CONCLUSIONS: This mobile-based intervention delivered by a multidisciplinary team can better improve glycemic control rates of patients with type 2 diabetes than usual care. These effects were best sustained within the first 6 months. Starting from the sixth month, intensive management needs to be conducted to maintain long-term effectiveness of the mobile-based intervention.

Prognostic Value of Hyperechoic Echo Halo in cN0 Papillary Thyroid Microcarcinoma and Its Correlation with Age and Gender.

Purpose. To investigate the pathology and prognostic value of hyperechoic echo halo in cN0 papillary thyroid microcarcinoma (PTMC) and the relationship between age, gender, and the formation of abnormal hyperechoic echo halo and cervical lymph node metastasis. Data of 97 patients who underwent surgical treatment for the first time for single PTMC between April 2016 and March 2017 were analyzed retrospectively. The boundary status of the PTMC was determined preoperatively. Grayscale values of the nodular center, hyperechoic echo halo, and normal thyroid tissue were acquired with Adobe Photoshop CS6 software. The histopathology of the boundary and status of the cervical lymph node metastasis were analyzed. Formation of abnormal hyperechoic halo and cervical lymph node metastasis in relation to age and gender were explored. The abnormal hyperechoic halo mainly represents cancer cell infiltration with reactive hyperplasia of inflammatory cells and fibrous tissue. In the hyperechoic halo group, the grayscale values for the nodular center, hyperechoic echo halo, and normal thyroid tissue were 1552.6 ± 578.6, 5792.0 ± 747.6, and 3582.7 ± 759.0, respectively (P < 0.05). The cervical lymph node metastasis rate was significantly lower in patients with hyperechoic halo (15.0%) than in those without (41.6%; P < 0.05) and significantly higher in those aged <45 years (53.3%) than in those aged ≥45 years (28.4%; P < 0.05). There were no significant correlations between gender and cervical lymph node metastasis or between age, gender, and hyperechoic halo formation (P > 0.05). cN0 PTMC patients with abnormal hyperechoic halo and age >45 years have a significantly reduced risk of cervical lymph node metastasis and relatively good prognosis.