Machine learning on MRI radiomic features: identification of molecular subtype alteration in breast cancer after neoadjuvant therapy.

OBJECTIVES: Recent studies have revealed the change of molecular subtypes in breast cancer (BC) after neoadjuvant therapy (NAT). This study aims to construct a non-invasive model for predicting molecular subtype alteration in breast cancer after NAT. METHODS: Eighty-two estrogen receptor (ER)-negative/ human epidermal growth factor receptor 2 (HER2)-negative or ER-low-positive/HER2-negative breast cancer patients who underwent NAT and completed baseline MRI were retrospectively recruited between July 2010 and November 2020. Subtype alteration was observed in 21 cases after NAT. A 2D-DenseUNet machine-learning model was built to perform automatic segmentation of breast cancer. 851 radiomic features were extracted from each MRI sequence (T2-weighted imaging, ADC, DCE, and contrast-enhanced T1-weighted imaging), both in the manual and auto-segmentation masks. All samples were divided into a training set (n = 66) and a test set (n = 16). XGBoost model with 5-fold cross-validation was performed to predict molecular subtype alterations in breast cancer patients after NAT. The predictive ability of these models was subsequently evaluated by the AUC of the ROC curve, sensitivity, and specificity. RESULTS: A model consisting of three radiomics features from the manual segmentation of multi-sequence MRI achieved favorable predictive efficacy in identifying molecular subtype alteration in BC after NAT (cross-validation set: AUC = 0.908, independent test set: AUC = 0.864); whereas an automatic segmentation approach of BC lesions on the DCE sequence produced good segmentation results (Dice similarity coefficient = 0.720). CONCLUSIONS: A machine learning model based on baseline MRI is proven useful for predicting molecular subtype alterations in breast cancer after NAT. KEY POINTS: • Machine learning models using MRI-based radiomics signature have the ability to predict molecular subtype alterations in breast cancer after neoadjuvant therapy, which subsequently affect treatment protocols. • The application of deep learning in the automatic segmentation of breast cancer lesions from MRI images shows the potential to replace manual segmentation..

Co-occurrence trend of school bullying and depression and its correlation with dietary patterns among middle school students from Inner Mongolia Autonomous Region in 2021 / 中国学校卫生

Objective@#To investigate the correlation between school bullying and depressive symptoms comorbidity and dietary patterns among middle school students in Inner Mongolia Autonomous Region from 2021, so as to provide reference for the prevention of school bullying and depressive symptoms.@*Methods@#In September 2021, stratified random cluster sampling was used to select 87 414 middle school students in 12 leagues in Inner Mongolia Autonomous Region. The Center for Epidemiologic Studies Depression Scale (CES-D) was used to assess depression, and the bullying was determined according to the items related to bullying in the program of Chinese National Surveillance on Students Common Diseases and Risk Factors.@*Results@#In 2021, the detection rate of depressive symptoms among middle school students in Inner Mongolia Autonomous Region was 15.24%, school bullying was 3.02%, and the co-occurrence of school bullying and depression was 1.64%. Binary Logistic regression analysis showed that junior high school students ( OR =1.52) and girls ( OR =1.10) were more likely to suffer from comorbidity of school bullying and depression ( P < 0.05). Eating fried food less than one and more than once a day, smoking and drinking were positively correlated with school bullying and depression comorbidity ( OR =2.15,2.11,2.14,1.70, P <0.05).@*Conclusion@#The combination of bullying and depression among middle school students in Inner Mongolia Autonomous Region is affected by various dietary methods. In terms of diet, reducing the intake of fried food, no smoking, no drinking can effectively reduce the incidence of co-occurrence school bullying and depression.

Prevalence trend and related factors of scoliosis and myopia co-morbidities among primary and middle school students in Inner Mongolia Autonomous Region from 2019 to 2022 / 中国学校卫生

Objective@#To analyze the prevalence trend of scoliosis among myopic students in Inner Mongolia Autonomous Region during 2019-2022, to explore the common etiology of myopia and scoliosis co-morbidities, so as to provide a reference for the development of relevant measures.@*Methods@#The method of stratified random cluster sampling was used to select 181 533, 141 552 , 200 987, 190 918 primary and secondary school students from 12 leagues(103 banners) in Inner Mongolia Atuonomous Region in September each year from 2019 to 2022. And scoliosis screening, vision examination and questionnaire survey were conducted among students in the included studies. The χ 2 test was used to analyze the data, and the binary multivariate Logistic regression model was used to screen the influencing factors of scoliosis and myopia co-morbidities.@*Results@#From 2019 to 2022, the myopia rate of primary and secondary school students in Inner Mongolia Autonomous Region was 55.55%-59.72%, scoliosis rate was 1.56 %-2.81% and the rates of scoliosis and myopia co-morbidities were 1.14%-1.95%, and the difference between different years was statistically significant ( χ 2=595.01, 775.56, 461.84, P < 0.05 ). In 2022, the co-morbidity rate was higher in girls than in boys(1.32% vs 0.97%), the rate of urban areas was higher than that of rural areas(2.57% vs 0.62%), the rate of students in vocational high school and high school was higher than that in junior high school and primary school (3.82%,2.47% vs 1.70%,0.42%), the rate of over developed areas was higher than that of poor areas (1.21% vs 0.99%)( χ 2=52.19, 1 269.82, 1 361.52, 17.29, P < 0.05 ). Logistic regression analysis showed that at least 1 h of moderate and high intensity exercise every day on weekends, the number of physical education classes per week was more than 3 sessions, the height of desks and chairs was adjusted according to height, resting outdoors, limiting screen time, and strictly requiring standing and sitting posture were the negative correlated with scoliosis and myopia, and the OR value was 0.65-0.90, reading books or electronic screens while participating in cram classes, walking or riding in the car were positively correlated with comorbid scoliosis and myopia, and the OR values were 1.27 and 1.13 ( P < 0.05), respectively.@*Conclusion@#Behavioral habits severely affect scoliosis and myopic of students. Prevention and control of scoliosis and myopia co-morbidity should start with students behavioral habits, early screening and early intervention.

Co-morbidity of overweight,obesity and depression and its influencing factors among students in Inner Mongolia Autonomous Region / 中国学校卫生

Objective@#To analyze the prevalence of overweight, obesity and depression among students in the Inner Mongolia Autonomous Region in 2019-2022 and explore the relevant factors affecting the co-morbidity of overweight,obesity and depression among students, so as to provide scientific basis for the prevention of co-morbidity.@*Methods@#From September 2019 to 2022, used stratified random cluster sampling, 90 519,71 560,90 079,91 089 students were selected from all 12 leagues in Inner Mongolia Autonomous Region for questionnaire survey and physical examination. The χ 2 test was used for demographic characteristics and univariate analysis, and a binary Logistic regression model was used to explore the association between lifestyle behaviors and co-morbidity.@*Results@#The detection rate of overweight,obesity among students from 2019-2022 was 29.21%,34.38%,35.20%,34.61%, the detection rate of depression was 18.35%，17.53%，16.43%，16.00%, and the co-morbidity detection rate of the two was 5.52%，5.93%，5.76%，5.46%. The number of overweight,obesity and depression co-morbidity students in 2022 was 4 978 students, and the co-morbidity detection rates of the students were significantly different in terms of the school segments and the family structures ( χ 2=103.51, 99.90, P <0.01). Multivariate Logistic regression analysis showed that consuming sugar sweetened beverages or fried food ≥1 time/d, sometimes or never eat breakfast, watching computer or TV ≥2 h/d, and less than 1 h of moderate-to-vigorous physical activity on weekends were positively correlated with the occurrence of the co-morbidity of overweight,obesity and depression, with the value of OR ranging from 1.17 to 1.59 ( P <0.05). Eated fresh fruits or drinking milk and soy milk ≥1 time/d, outdoor activities ≥1 h/d, sleep ≥8 h/d, not smoking and not drinking alcohol were negatively correlated with the occurrence of comorbid overweight,obesity and depression, with the value of OR ranging from 0.47 to 0.92 ( P < 0.05).@*Conclusion@#The occurrence of overweight,obesity and depressive symptoms co-morbidity in students is associated with dietary, exercise and lifestyle behavior. Targeted measures should be taken to maintain students healthy weight and prevent the occurrence of depression from the aspects of diet, exercise and life habits.