The prevalence of neurodevelopment disorders (NDDs) among children has been on the rise. This has affected the health and social life of children. This condition has also imposed a huge economic burden on families and health care systems. Currently, it is difficult to perform early diagnosis of NDDs, which results in delayed intervention. For this reason, patients with NDDs have a prognosis. In recent years, machine learning (ML) technology, which integrates artificial intelligence technology and medicine, has been applied in the early detection and prediction of diseases based on data mining. This paper reviews the progress made in the application of ML in the diagnosis and treatment of NDDs in children based on supervised and unsupervised learning tools. The data reviewed here provide new perspectives on early diagnosis and treatment of NDDs.
Early detection of children with autism spectrum disorder (ASD) and comorbid intellectual disability (ID) can help in individualized intervention. Appropriate assessment and diagnostic tools are lacking in primary care. This study aims to explore the applicability of machine learning (ML) methods in diagnosing ASD comorbid ID compared with traditional regression models.From January 2017 to December 2021, 241 children with ASD, with an average age of 6.41 ± 1.96, diagnosed in the Developmental Behavior Department of the Children's Hospital Affiliated with the Medical College of Zhejiang University were included in the analysis. This study trained the traditional diagnostic models of Logistic regression (LR), Support Vector Machine (SVM), and two ensemble learning algorithms [Random Forest (RF) and XGBoost]. Socio-demographic and behavioral observation data were used to distinguish whether autistic children had combined ID. The hyperparameters adjustment uses grid search and 10-fold validation. The Boruta method is used to select variables. The model's performance was evaluated using discrimination, calibration, and decision curve analysis (DCA).Among 241 autistic children, 98 (40.66%) were ASD comorbid ID. The four diagnostic models can better distinguish whether autistic children are complicated with ID, and the accuracy of SVM is the highest (0.836); SVM and XGBoost have better accuracy (0.800, 0.838); LR has the best sensitivity (0.939), followed by SVM (0.952). Regarding specificity, SVM, RF, and XGBoost performed significantly higher than LR (0.355). The AUC of ML (SVM, 0.835 [95% CI: 0.747-0.944]; RF, 0.829 [95% CI: 0.738-0.920]; XGBoost, 0.845 [95% CI: 0.734-0.937]) is not different from traditional LR (0.858 [95% CI: 0.770-0.944]). Only SVM observed a good calibration degree. Regarding DCA, LR, and SVM have higher benefits in a wider threshold range.Compared to the traditional regression model, ML model based on socio-demographic and behavioral observation data, especially SVM, has a better ability to distinguish whether autistic children are combined with ID.
Accurate prediction of energy requirement is important in formulating diets, but an energy model for Yellow Broiler breeder hens is publicly unavailable. The objective of this study was to establish energy prediction models for the nitrogen-corrected apparent metabolisable energy (AMEn) requirement of different categories of Yellow Broiler breeder hens during the egg-laying period. Data for modelling were collected from research papers, public databases and production data from companies. Breeder hens were generally categorised into three BW types: heavy, medium and light (HBWT, MBWT and LBWT). Published articles were cited for providing coefficients of AMEn maintenance requirement (AMEnm, 101 kcal/kg BW0.75, 423 KJ/kg BW0.75) and growth requirement (AMEng, 5.33 kcal/g, 22.3 KJ/g), respectively. Models of AMEn for egg production (AMEnp) were established from the known daily intake of AMEn (AMEni) and those of maintenance and growth by the factorial approach: AMEnp = AMEni - AMEnm - AMEng. For the three types of hens, AMEnp HBWT (kcal, KJ) = 2.55 kcal (10.7 KJ) × egg mass (EM, g); AMEnp MBWT (kcal, KJ) = 2.70 kcal (11.3 KJ) × EM (g), and AMEnp LBWT (kcal, KJ) = 2.94 kcal (12.3 KJ) × EM (g) were determined. The total AMEni requirements, depending on Gompertz models, were HBWT: BW (g) = 3 144 × e-EXP(-0.162×(week of age (wk)-15.6)); MBWT: BW (g) = 2 526 × e-EXP(-0.333×(wk-19.1)); LBWT: BW (g) = 1 612 × e-EXP(-0.242×(wk-16.5)). Models of egg production, HBWT: egg production (%) = 124 × e-0.017×wk/(1 + e-0.870×(wk-26.2)); MBWT: egg production (%) = 144 × e-0.020×wk/(1 + e-0.751×(wk-24.9)); LBWT: egg production (%) = 163 × e-0.024×wk/(1 + e-0.476×(wk-26.5))) and egg weight for each wk of the three types of hens during the egg-laying period were all established. These models showed good applicability in simulating and predicting the literature or production data.
The staining procedure is critical for investigating intra- and extra-cellular ultrastructure of microorganisms by transmission electron microscopy (TEM). Here, we propose a new ultra-low lead staining (ULLS) technique for preparing the ultrathin sections for TEM analysis. Sections of Enterobacter sp. (bacteria), Aspergillus niger (filamentous fungi), Rhodotorula mucilaginosa (fungi), and Chlamydomonas reinhardtii (microalgae) were tested. Compared with the sections prepared by the typical double-staining technique, ULLS-based sections showed evident advantages: (i) the staining process only required the addition of Pb(NO 3 ) 2 ; (ii) the Pb level during incubation was set as low as 1 mg/L, which had negligible toxicity to most microbial cells; (iii) the Pb cations were added during microbial culture, which avoided complicated sample preparation as in typical double staining. Taking C. reinhardtii as an example, the ULLS technique allowed fine investigation of microbial ultrastructure, e.g., starch granule, mitochondrion, Golgi apparatus, vacuole, and vesicle. Meanwhile, the physiological processes of the cells such as cell lysis and exocytosis were successfully captured, with relatively high contrast. This study hence shows a bright future on preparation of the high-quality ultrathin sections of microbial cells by the ULLS technique.
Abstract Background : Prediction of stroke based on individuals’ risk factors, especially for a first stroke event, is of great significance for primary prevention of high-risk populations. Our study aimed to investigate the applicability of interpretable machine learning for predicting a 2-year stroke occurrence in older adults compared with logistic regression. Methods : A total of 5960 participants consecutively surveyed from July 2011 to August 2013 in the China Health and Retirement Longitudinal Study were included for analysis. We constructed a traditional logistic regression (LR) and two machine learning methods, namely random forest (RF) and extreme gradient boosting (XGBoost), to distinguish stroke occurrence versus non-stroke occurrence using data on demographics, lifestyle, disease history, and clinical variables. Grid search and 10-fold cross validation were used to tune the hyperparameters. Model performance was assessed by discrimination, calibration, decision curve and predictiveness curve analysis. Results : Among the 5960 participants, 131 (2.20%) of them developed stroke after an average of 2-year follow-up. Our prediction models distinguished stroke occurrence versus non-stroke occurrence with excellent performance. The AUCs of machine learning methods (RF, 0.823[95% CI, 0.759-0.886]; XGBoost, 0.808[95% CI, 0.730-0.886]) were significantly higher than LR (0.718[95% CI, 0.649, 0.787], p <0.05). No significant difference wa s observed between RF and XGBoost ( p >0.05). All prediction models had good calibration results, and the brier score were 0.022 (95% CI, 0.015-0.028) in LR, 0.019 (95% CI, 0.014-0.025) in RF, and 0.020 (95% CI, 0.015-0.026) in XGBoost. XGBoost had much higher net benefits within a wider threshold range in terms of decision curve analysis, and more capable of recognizing high risk individuals in terms of predictiveness curve analysis. A total of eight predictors including gender, waist-to-height ratio, dyslipidemia, glycated hemoglobin, white blood cell count, blood glucose, triglycerides, and low-density lipoprotein cholesterol ranked top 5 in three prediction models. Conclusions : Machine learning methods, especially for XGBoost, had the potential to predict stroke occurrence compared with traditional logistic regression in the older adults.
A 5 × 2 × 3 factorial experiment was used to investigate the effects of 5 ME levels (12.55, 12.30, 12.05, 11.80, and 11.55 MJ/kg) supplemented with or without exogenous enzymes in diets of broilers on the nutrient digestibility and energy improving efficiency over the starter, grower, and finisher phases of growth. The results indicated that the apparent digestibility of DM decreased linearly with a reduction in the ME level in diets for the starter (R2 = 0.234, P < 0.001) and grower (R2 = 0.362, P < 0.001) phases, and increased with enzyme supplementation for all diets. The greatest improvement occurred in the diet with the lowest ME level. The AME value also decreased linearly with the reduction of ME level in diets (R2 = 0.418, P < 0.001 for starter; R2 = 0.398, P < 0.001 for grower; R2 = 0.097, P = 0.027 for finisher). Enzyme supplementation enhanced the AME value of diets in the starter, grower, and finisher phases by 0.07 ~ 0.62, 0.15 ~ 0.56, and 0.12 ~ 0.43 MJ/kg, respectively, and the optimal improvement of AME value occurred when the ME of diet was 11.55 MJ/kg in the starter phase. The effects of enzyme addition on AME for the starter phase were significantly greater than for the other phases. A significant interaction between ME level and enzyme supplementation in growth stage (P < 0.05) was observed. The retention of CP decreased linearly with the reduction of ME level in diets (R2 = 0.245, P < 0.001 for starter; R2 = 0.367, P < 0.001 for grower). The retention of CP was increased by enzyme supplemented into the diets with ME levels of 11.55 and 11.80 MJ/ kg. Together, our results suggested that the ME level of diet affected the digestibility of DM, energy, and CP, and enzyme supplementation improved energy digestibility in diets with lower levels of ME.
Although multimorbidity is a risk factor for disability, the relationship between the accumulative patterns of multimorbidity and disability remains poorly understood. The objective of this study was to identify the latent groups of multimorbidity trajectories among mid to older age adults and to examine their associations with incident disability.We included 5,548 participants aged ≥ 45 years who participated in the China Health and Retirement Longitudinal Study from 2011 to 2018 and had no multimorbidity (≥ 2 chronic conditions) at baseline. The group-based multi-trajectory modeling was used to identify distinct trajectory groups of multimorbidity based on the latent dimensions underlying 13 chronic conditions. The association between multimorbidity trajectories and incident disability was analyzed using the generalized estimating equation model adjusting for potential confounders.Of the 5,548 participants included in the current analysis, 2,407 (43.39%) developed multimorbidity during the follow-up. Among participants with new-onset multimorbidity, four trajectory groups were identified according to the combination of newly diagnosed diseases: "Cardiometabolic" (N = 821, 34.11%), "Digestive-arthritic" (N = 753, 31.28%), "Cardiometabolic/Brain" (N = 618, 25.68%), and "Respiratory" (N = 215, 8.93%). Compared to participants who did not develop multimorbidity, the risk of incident disability was most significantly increased in the "Cardiometabolic/Brain" trajectory group (OR = 2.05, 95% CI: 1.55-2.70), followed by the "Cardiometabolic" (OR = 1.96, 95% CI: 1.52 -2.53) and "Digestive-arthritic" (OR = 1.70, 95% CI: 1.31-2.20) trajectory groups.The growing burden of multimorbidity, especially the comorbid of cardiometabolic and brain diseases, may be associated with a significantly increased risk of disability for mid to older age adults. These findings improve our understanding of multimorbidity patterns that affect the independence of living and inform the development of strategies for the primary prevention of disability.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)