Objective To construct an artificial intelligence follow-up system for multiple myeloma patients based on patient needs，and to evaluate its application effect. Methods From March to May 2022，308 patients with multiple myeloma were selected from 3 tertiary hospitals in Shandong via convenience sampling method，and they were investigated by the cognition and demand content questionnaire of the artificial intelligence follow-up system，and the module of the follow-up system was developed，and its functions were combined with the initial application of the follow-up system for patients with hematologic diseases. From July to December 2022，patients’ symptom management level and satisfaction before and after application of the system were evaluated by the cancer patient self-management assessment scale and the continuous Care Smart platform using the questionnaire. The continuous Care Smart platform was used to evaluate the application satisfaction of medical staff with the questionnaire. Results The artificial intelligence follow-up system for multiple myeloma patients consisted of 2 modules and 9 functions. After the system was applied，the self-management level of 83 patients with multiple myeloma was higher than that before application（P<0.05）；the satisfaction score of patients was （67.36±8.27） points；the timely completion rate of follow-up was higher than that before the application （P<0.01）. The satisfaction of medical staff on the practicability，convenience，reliability and accuracy of the system was more than 95%，and the satisfaction of intelligence was 84.6%. Conclusion The artificial intelligence follow-up system for multiple myeloma patients has a scientific construction process，good practicability and convenience，and patients have high satisfaction with it.

Objective To translate the English version of the Supportive Needs of Breastfeeding Scale（SNB） during hospitalization into Chinese,and to verify its reliability and validity. Methods The original scale was translated,back-translated,culturally adapted using the Brislin translation model to form the Chinese version of the scale. A convenience sampling method was used to survey 360 mothers in a tertiary specialized hospital in Shanghai from March to July 2021,in order to analyze the reliability and validity of the scale. Results The Chinese version of the scale included 4 dimensions of emotional needs,skill needs,informational needs and appraisal needs,with a total of 18 items. The content validity of the scale I-CVI content validity index ranged from 0.830 to 1.000,and the S-CVI mean content validity index was 0.930. Exploratory factor analysis used principal component analysis to extract 4 common factors with a cumulative variance contribution rate of 76.152%. After confirmatory factor analysis, χ²/df was 2.568;RMSEA was 0.023;the equations fit well. The Chinese version of the Breastfeeding Self-Efficacy Scale Short-Form was used as the calibration standard,and the calibration standard correlation validity was 0.714. The Cronbach’s alpha coefficient of the scale was 0.947;the split-half reliability was 0.953,and the test-retest reliability was 0.912. Conclusion The scale has good reliability and validity,and it can be used as an assessment tool for breastfeeding support needs during hospitalization.

Objective To systematically evaluate the application effect of prone ventilation in patients with hypoxemia after cardiopulmonary bypass surgery，so as to provide a reference for further discussing the effective mechanical ventilation position of such patients. Methods Pubmed，Web of Science，Embase，Cochrane Library，CENTRAL，CBM，CNKI，Wanfang Data and VIP database were searched from the establishment of databases until September，2021. All the published randomized controlled trials（RCTs） about the effects of prone position ventilation on patients with hypoxemia after cardiac surgery were confirmed by the literature reference citation retrieval. Inclusion and exclusion criteria were used to review and rank the literature. After the evaluation of the quality of included studies，the data was extracted from RCTs and a meta-analysis was conducted by RevMen 5.4. Results 9 RCTs with 456 cases were included. The results of meta-analysis showed that in oxygenation index，the oxygenation index of prone ventilation was higher than that of supine ventilation when the ventilation time（T1） in prone position≥12 h，and the start time（T2） was 6 hours after surgery and the intermittent prone ventilation，and the difference was statistically significant [MD=57.31，95%CI（35.84，78.77），P<0.001]，[MD=71.98，95%CI（64.34，79.61），P<0.001]，[MD=37.97，95%CI（18.36，57.58），P=0.001]. In oxygen partial pressure，when the oxygen concentration（FiO₂） of prone ventilation was lower than that of supine ventilation，the oxygenation index of prone ventilation was higher than that of supine ventilation，and the difference was statistically significant[MD=5.70，95%CI（0.46，10.94），P=0.03]. In clinical outcome，the mechanical ventilation time[MD=-56.47，95%CI（-72.62，-40.33），P<0.001]，ICU stay time[MD=-60.82，95%CI（-76.82，-44.81），P<0.001] and hospital stay time[MD=-44.36，95%CI（-63.10，-25.62），P<0.001] in prone ventilation group were lower than those in supine ventilation group，and the differences were statistically significant. Conclusion The prone position ventilation is effective to improve PaO₂/FiO₂，duration of mechanical ventilation，duration of ICU care，and hospital stays in patients with hypoxemia after cardiac surgery，while it could not effectively improve PaO₂. Despite subgroup analysis of the main indicators，the optimal duration and initiation of prone ventilation remained unclear. Due to the limitations of the quality of included studies，randomized controlled trials with large sample size are needed in the future to evaluate the effects of prone position ventilation on neonatal respiratory distress syndrome.