Childhood Leukemia Classification via Information Bottleneck Enhanced Hierarchical Multi-Instance Learning.

Leukemia classification relies on a detailed cytomorphological examination of Bone Marrow (BM) smear. However, applying existing deep-learning methods to it is facing two significant limitations. Firstly, these methods require large-scale datasets with expert annotations at the cell level for good results and typically suffer from poor generalization. Secondly, they simply treat the BM cytomorphological examination as a multi-class cell classification task, thus failing to exploit the correlation among leukemia subtypes over different hierarchies. Therefore, BM cytomorphological estimation as a time-consuming and repetitive process still needs to be done manually by experienced cytologists. Recently, Multi-Instance Learning (MIL) has achieved much progress in data-efficient medical image processing, which only requires patient-level labels (which can be extracted from the clinical reports). In this paper, we propose a hierarchical MIL framework and equip it with Information Bottleneck (IB) to tackle the above limitations. First, to handle the patient-level label, our hierarchical MIL framework uses attention-based learning to identify cells with high diagnostic values for leukemia classification in different hierarchies. Then, following the information bottleneck principle, we propose a hierarchical IB to constrain and refine the representations of different hierarchies for better accuracy and generalization. By applying our framework to a large-scale childhood acute leukemia dataset with corresponding BM smear images and clinical reports, we show that it can identify diagnostic-related cells without the need for cell-level annotations and outperforms other comparison methods. Furthermore, the evaluation conducted on an independent test cohort demonstrates the high generalizability of our framework.

The Associations of Caesarean Delivery With Risk of Wheezing Diseases and Changes of T Cells in Children.

Objectives: This study aimed to assess the associations of caesarean delivery (CD) with risk of wheezing diseases and changes of immune cells in children. Design: The cross-sectional study was conducted between May, 2020 and April, 2021. Setting and participants: The study was conducted in Shanghai Children's Medical Center, Shanghai, China. A total of 2079 children with a mean age of 36.97 ± 40.27 months and their guardians were included in the present study via face-to-face inquiry and physical examination by clinicians. Methods: Logistic regression was applied to estimate odds ratio (ORs) and 95% confidence intervals (CIs) for the association between CD and first episode of wheezing (FEW) or asthma. Models were adjusted for premature or full-term delivery, exclusive breastfeeding (at least 4 months) or not. Results: Among the 2079 children, 987 children (47.47%) were born by CD and 1092 (52.53%) by vaginal delivery (VD). Children delivered by caesarean had significantly lower gestational age (P<0.01) compared with those who delivered vaginally. Our results also showed that CD was related to increased risk of FEW by the age of 3(adjusted OR 1.50, 95%CI 1.06, 2.12) and increased tendency to develop asthma by the age of 4 (adjusted OR 3.16, 95%CI 1.25, 9.01). The subgroup analysis revealed that the negative effects of CD on asthma were more obvious in children without exclusive breastfeeding (adjusted OR 4.93, 95%CI 1.53, 21.96) or without postnatal smoking exposure (adjusted OR 3.58, 95%CI 1.20, 13.13). Furthermore, compared with children born through VD, a significant change of the T cells (increased proportion of CD4+ T cells and decreased number and proportion of CD8+ T cells) were observed before the age of one in the CD group. However, the changes were insignificant in children over 1 year old. Conclusions: This study showed age-dependent associations of CD with asthma and FEW in offspring. Moreover, CD appeared to have an effect on the cellular immunity in infants, the disorder of which may contribute to the development of asthma in children.

Machine Learning for Early Warning of Septic Shock in Children With Hematological Malignancies Accompanied by Fever or Neutropenia: A Single Center Retrospective Study.

OBJECTIVES: The purpose of this article was to establish and validate clinically applicable septic shock early warning model (SSEW model) that can identify septic shock in hospitalized children with onco-hematological malignancies accompanied with fever or neutropenia. METHODS: Data from EMRs were collected from hospitalized pediatric patients with hematological and oncological disease at Shanghai Children's Medical Center. Medical records of patients (>30 days and <19 years old) with fever (≥38°C) or absolute neutrophil count (ANC) below 1.0 × 109/L hospitalized with hematological or oncological disease between January 1, 2017 and August 1, 2019 were considered. Patients in whom septic shock was diagnosed during the observation period formed the septic shock group, whereas non-septic-shock group was the control group. In the septic shock group, the time points at 4, 8, 12, and 24 hours prior to septic shock were taken as observation points, and corresponding observation points were obtained in the control group after matching. We employed machine learning artificial intelligence (AI) to filter features and used XGBoost algorithm to build SSEW model. Area under the ROC curve (AU-ROC) was used to compare the effectiveness among the SSEW Model, logistic regression model, and pediatric sequential organ failure score (pSOFA) for early warning of septic shock. MAIN RESULTS: A total of 64 observation periods in the septic shock group and 2191 in the control group were included. AU-ROC of the SSEW model had higher predictive value for septic shock compared with the pSOFA score (0.93 vs. 0.76, Z = -2.73, P = 0.006). Further analysis showed that the AU-ROC of the SSEW model was superior to the pSOFA score at the observation points 4, 8, 12, and 24 h before septic shock. At the 24 h observation point, the SSEW model incorporated 14 module root features and 23 derived features. CONCLUSION: The SSEW model for hematological or oncological pediatric patients could help clinicians to predict the risk of septic shock in patients with fever or neutropenia 24 h in advance. Further prospective studies on clinical application scenarios are needed to determine the clinical utility of this AI model.

Development and Evaluation of a Leukemia Diagnosis System Using Deep Learning in Real Clinical Scenarios.

Leukemia is the most common malignancy affecting children. The morphologic analysis of bone marrow smears is an important initial step for diagnosis. Recent publications demonstrated that artificial intelligence is able to classify blood cells but a long way from clinical use. A total of 1,732 bone marrow images were used for the training of a convolutional neural network (CNN). New techniques of deep learning were integrated and an end-to-end leukemia diagnosis system was developed by using raw images without pre-processing. The system creatively imitated the workflow of a hematologist by detecting and excluding uncountable and crushed cells, then classifying and counting the remain cells to make a diagnosis. The performance of the CNN in classifying WBCs achieved an accuracy of 82.93%, precision of 86.07% and F1 score of 82.02%. And the performance in diagnosing acute lymphoid leukemia achieved an accuracy of 89%, sensitivity of 86% and specificity of 95%. The system also performs well at detecting the bone marrow metastasis of lymphoma and neuroblastoma, achieving an average accuracy of 82.93%. This is the first study which included a wider variety of cell types in leukemia diagnosis, and achieved a relatively high performance in real clinical scenarios.

Association between smoking and glycemic control in diabetic patients: Results from the Risk Evaluation of cAncers in Chinese diabeTic Individuals: A lONgitudinal (REACTION) study.

BACKGROUND: A number of primary studies suggested that active smoking could be independently associated with incident diabetes. However less is known about the effect of active smoking and smoking cessation on glycemic control in patients with diabetes. The aim of this study was to evaluate the associations of active smoking and smoking cessation with glycemic control in diabetic patients. METHODS: The present was a cross-sectional study of 10 551 men and 15 297 women with diabetes from the Risk Evaluation of cAncers in Chinese diabeTic Individuals: a lONgitudinal (REACTION) study. Risk factors for glycemic control and the association of active smoking with glycemic control were evaluated using logistic regression models. Poor glycemic control was defined as HbA1c ≥7.0%. RESULTS: Current smokers have an increased risk of poor glycemic control, and the multivariable-adjusted odds ratio (OR) and 95% confidence intervals (CI) of HbA1c ≥7.0% with current smoking were 1.49 (1.35-1.66) in men and 1.56 (1.13-2.15) in women. Further analysis demonstrated a dose-dependent relationship between active smoking and the risk of poor glycemic control in men. Former smokers who quit smoking for <10 years remained at increased risk of poor glycemic control, with the risk leveling off after 10 years of smoking cessation compared with non-smokers, but risk in former smokers was significantly lower than that in current smokers. CONCLUSIONS: Active smoking is a modifiable risk factor for poor glycemic control in Chinese diabetic patients.

Glycemic status and chronic kidney disease in Chinese adults: Findings from the REACTION study.

BACKGROUND: Diagnosed diabetes has been associated with chronic kidney disease (CKD). However, the association between non-diabetic hyperglycemia and CKD remained uncertain. The aim of the present study was to investigate the association between different glycemic status and CKD in Chinese adults and to assess the prevalence and control of diabetes among individuals with CKD. METHODS: In all, 250 752 adults aged ≥40 years were selected from the baseline cohort of the Risk Evaluation of cAncers in Chinese diabeTic Individuals: a lONgitudinal (REACTION) study. Plasma glucose concentrations and biochemical and other clinical data were collected; CKD was defined as an estimated glomerular filtration rate (eGFR) <60 mL/min per 1.73 m2 . RESULTS: The prevalence of CKD increased gradually with deterioration of glucose metabolic status in both men and women ( P trend < 0.001 for both). Compared with individuals with normal glucose regulation, men with prediabetes and diabetes had higher risks of prevalent CKD (prediabetes odds ratio [OR] 1.15, 95% confidence interval [CI] 1.02-1.32; newly diagnosed diabetes OR 1.27, 95% CI 1.08-1.49; previously diagnosed diabetes OR 2.05, 95% CI 1.78-2.35). Similar results were observed in women, but not among those with prediabetes. In male CKD patients with diabetes, 52.1% received antidiabetic treatment and 41.8% of those treated had effective glycemic control, higher than values for females. CONCLUSIONS: Prediabetes and diabetes were associated with an increased risk of CKD in Chinese men. Control of diabetes among Chinese CKD patients is far from optimal.

Association of insulin resistance with breast, ovarian, endometrial and cervical cancers in non-diabetic women.

Hyperinsulinemia and insulin resistance were reported to play a crucial role in diabetes-cancer relationship. This study aimed to explore the associations between insulin resistance and several female cancers in a non-diabetic population. This cross-sectional study was conducted in 121,230 middle-aged and elderly non-diabetic women. Cancer diagnosis was self-reported and further validated by medical records. Insulin resistance was defined as homeostasis model assessment of insulin resistance (HOMA-IR) ≥ 2.50. The prevalence of both premenopausal and postmenopausal breast cancer, postmenopausal ovarian cancer and premenopausal endometrial cancer were higher in insulin-resistant participants than in insulin-sensitive participants (premenopausal breast cancer, 0.45 vs 0.28%; postmenopausal breast cancer, 0.86 vs 0.63%; postmenopausal ovarian cancer, 0.17 vs 0.09%; premenopausal endometrial cancer, 0.43 vs 0.25%, respectively, all P < 0.05). Individuals with insulin resistance had higher odds ratio (OR) of breast cancer, both premenopausal and postmenopausal (OR 1.98, 95% confidence interval (CI) 1.19-3.32; OR 1.29, 95% CI 1.01-1.63), postmenopausal ovarian cancer (OR 2.17, 95% CI 1.10-3.40) as well as total endometrial cancer (OR 1.47, 95% CI 1.02-2.12). Subgroup analysis revealed that the possitive association between insulin resistance and risk of prevalent breast cancer was observed in popualtion with younger age, overweight or obesity, higher education and impaired glucose tolerance (IGR). No relationships were observed for the risk of prevalent cervical cancers with insulin resistance. Non-diabetic women with insulin resistance had higher risk of prevalent breast, ovarian and endomatrial cancer, which suggests special attentions to these female cancer screening and prevention.

Prevalence of Diabetes and Cardiometabolic Disorders in Spouses of Diabetic Individuals.

Pairs of spouses share common lifestyle factors. In a cross-sectional analysis, we investigated whether spouses of diabetic individuals had a higher prevalence of diabetes and cardiometabolic disorders in a community-based population of Chinese adults aged 40 years or older between 2011 and 2012. A total of 34,805 pairs of spouses were identified. All participants underwent a standard oral glucose tolerance test and provided detailed clinical, sociodemographic, and lifestyle information. Diabetes and multiple cardiometabolic disorders were defined according to standard criteria. Compared with participants whose spouses did not have diabetes, participants whose spouses had diabetes had higher odds of having diabetes (for men, odds ratio (OR) = 1.33, 95% confidence interval (CI): 1.22, 1.45; for women, OR = 1.35, 95% CI: 1.24, 1.47), obesity (for men, OR = 1.34, 95% CI: 1.13, 1.59; for women, OR = 1.19, 95% CI: 1.05, 1.35), metabolic syndrome (for men, OR = 1.31, 95% CI: 1.21, 1.42; for women, OR = 1.12, 95% CI: 1.04, 1.20), and cardiovascular disease (for men, OR = 1.18, 95% CI: 1.03, 1.34; for women, OR = 1.18, 95% CI: 1.03, 1.35). The associations were independent of age, body mass index, education, family history of diabetes, cigarette smoking, alcohol drinking, physical activity, and diet. Spousal diabetes was simple and valuable information for identifying individuals at risk for diabetes and cardiometabolic disorders.

Reduced Kidney Function Is Associated With Cardiometabolic Risk Factors, Prevalent and Predicted Risk of Cardiovascular Disease in Chinese Adults: Results From the REACTION Study.

BACKGROUND: Chronic kidney disease (CKD) increases cardiovascular disease (CVD) risk. However, the association of mildly reduced kidney function with CVD risk is unclear. METHODS AND RESULTS: This study investigated the association of estimated glomerular filtration rate (eGFR) with prevalent CVDs, 10-year Framingham risk for coronary heart disease (CHD), and 10-year risk of atherosclerotic cardiovascular diseases (ASCVD) in 239 832 participants from the baseline of the Risk Evaluation of cAncers in Chinese diabeTic Individuals: a lONgitudinal study. With an interviewer-assisted questionnaire, we collected information on CVD, including reported CHD, stroke, or myocardial infarction. Chronic Kidney Disease-Epidemiology Collaboration (CKD-EPI) equation was used to calculate eGFR. Compared with individuals with normal eGFR (≥90 mL/min per 1.73 m(2)), those with decreased eGFR (75-89, 60-74, and <60 mL/min per 1.73 m(2)) had higher risk of prevalent obesity, diabetes mellitus, hypertension, and dyslipidemia in both men and women (P for trend all <0.001). Moreover, a significantly higher 10-year Framingham risk for CHD and 10-year risk for ASCVD was observed in both men and women with mildly decreased eGFR (60-89 mL/min per 1.73 m(2)). CONCLUSIONS: Even mildly reduced eGFR (under 90 mL/min per 1.73 m(2)) is associated with elevated 10-year Framingham risk for CHD and 10-year ASCVD risk among Chinese adults.

Risk of all-cause mortality and cardiovascular disease associated with secondhand smoke exposure: a systematic review and meta-analysis.

BACKGROUND: Emerging studies have assessed the association between secondhand smoke (SHS) exposure and cardiovascular disease (CVD) as well as all-cause mortality. However, findings were not consistent due to the heterogeneity of study characteristics. METHODS: PubMed and Embase were searched through May 2014 for prospective cohort and case-control studies investigating the associations of SHS exposure in never smokers with all-cause mortality and the risk of CVD. The main analysis was performed in studies using self-reported SHS exposure and secondary analysis was performed in studies using objectively measured SHS exposure. Summary estimates were calculated using random-effects models. RESULTS: Twenty-three prospective and 17 case-control studies were included. The pooled relative risks (RR) for never smokers exposed to SHS in comparison with those unexposed were 1.18 [95% confidence interval (CI): 1.10-1.27] for all-cause mortality (12 studies), and 1.23 (1.16-1.31) for CVD (38 studies). The association of SHS exposure with CVD was markedly stronger among studies conducted in China (RR=1.65, 95% CI 1.27-2.13) than that in the US (RR=1.09, 95% CI 1.03-1.16). Studies using objectively measured SHS exposure demonstrated a slightly higher risk for CVD compared with those using self-reported SHS exposure. CONCLUSIONS: Exposure to SHS significantly increased the risk for all-cause mortality and CVD. The risk associated with SHS exposure was large in China while the risk was only modest in the US. Studies using objectively measured SHS exposure may yield a higher risk of CVD than those using self-reported SHS exposure.

Cohort profile: risk evaluation of cancers in Chinese diabetic individuals: a longitudinal (REACTION) study.

OBJECTIVE: To demonstrate whether abnormal glucose metabolism (diabetes and prediabetes) is associated with increased risk for cancer in the Chinese population and to identify factors that modify the risk of cancer among individuals with abnormal glucose metabolism. METHODS: Between 2011 and 2012, 259 657 community-dwelling adults, aged 40 years and older, were recruited from 25 centers across mainland China to participant in the baseline survey of the REACTION study, with follow-up investigations performed 3, 5, and 10 years later. Detailed questionnaires about lifestyles, physical and biochemical measurement, bio-samples including serum, urine, and whole blood for DNA extraction were collected for all the participants. RESULTS: The mean ± standard deviation (SD) age of this cohort was 57 ± 10 years. And the prevalence of pre-existing and newly diagnosed diabetes was 10.32% and 10.57%, respectively. A total of 4511 prevalent cancer cases (988 men and 3523 women) were identified, the prevalence was 1.79. Compared to those with normal glucose metabolism, men with diabetes had a significantly lower adjusted prevalence ratio (PR) of stomach cancer (PR: 0.38, 95% CI: 0.16-0.89), and women with diabetes had significantly higher adjusted PRs of cancer of all sites (PR: 1.36, 95% CI: 1.20-1.56), and cancer of the breast (PR: 1.56, 95% CI: 1.21-2.00), the endometrium (PR: 1.58, 95% CI: 1.16-2.15), and the thyroid (PR: 1.53, 95% CI: 1.03-2.27). CONCLUSION: The multi-center REACTION study has captured a broad range of data on physical, psychological and metabolic function as well as health status, biochemical and lifestyle information in 259 657 adults from the general population across the China.

Association between alcohol consumption and metabolic syndrome in 19,215 middle-aged and elderly Chinese.

AIMS: To assess the association of alcohol consumption with metabolic syndrome (Mets) in a Chinese population. METHODS: A community-based cross-sectional study was performed in 19,215 participants aged 40 years or over in Shanghai of China during June 2008-April 2009. Information about the alcohol consumption including quantity and type of alcoholic beverage was obtained using a standard questionnaire. Physical examination was performed and fasting blood samples were obtained for the measurements of biochemical indicators. RESULTS: In male wine-only consumers, after adjusted for age, sex, BMI, education levels, exercise and smoking habit, severe alcohol consumption (≥ 50.0 g/d), compared with non-alcohol consumption, conferred 53% increased risk of having MetS. In women, alcohol consumption did not have relation to the prevalence of Mets. In the beer-only, liquor-only and mixed type consumers, no significant associations between MetS or its components and the quantity of alcohol consumption were detected. CONCLUSIONS: Our findings suggest that excessive wine consumption (alcohol ≥ 50.0 g/d) is associated with higher prevalence of MetS in Chinese men.