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.

Artificial intelligence promotes shared decision-making through recommending tests to febrile pediatric outpatients.

BACKGROUND: To promote the shared decision-making (SDM) between patients and doctors in pediatric outpatient departments, this study was designed to validate artificial intelligence (AI) -initiated medical tests for children with fever. METHODS: We designed an AI model, named Xiaoyi, to suggest necessary tests for a febrile child before visiting a pediatric outpatient clinic. We calculated the sensitivity, specificity, and F1 score to evaluate the efficacy of Xiaoyi's recommendations. The patients were divided into the rejection and acceptance groups. Then we analyzed the rejected examination items in order to obtain the corresponding reasons. RESULTS: We recruited a total of 11,867 children with fever who had used Xiaoyi in outpatient clinics. The recommended examinations given by Xiaoyi for 10,636 (89.6%) patients were qualified. The average F1 score reached 0.94. A total of 58.4% of the patients accepted Xiaoyi's suggestions (acceptance group), and 41.6% refused (rejection group). Imaging examinations were rejected by most patients (46.7%). The tests being time-consuming were rejected by 2,133 patients (43.2%), including rejecting pathogen studies in 1,347 patients (68.5%) and image studies in 732 patients (31.8%). The difficulty of sampling was the main reason for rejecting routine tests (41.9%). CONCLUSION: Our model has high accuracy and acceptability in recommending medical tests to febrile pediatric patients, and is worth promoting in facilitating SDM.

Racial differences in the associations of urinary phthalate metabolites with depression risk.

OBJECTIVE: This study aimed to investigate the composite effects of different kinds of phthalates on depression risk in the U.S population. METHODS: 11731 participants were included from the National Health and Nutrition Examination Survey (NHANES), a national cross-sectional survey. Twelve urinary phthalate metabolites were used to evaluate the level of phthalates exposure. Phthalates levels were devided into four quartiles. High phthalate was defined as having values in the highest quartile. RESULTS: Urinary mono-isobutyl phthalate (MiBP) and mono-benzyl phthalate (MBzP) were estimated as the independent risk factors for depression by mutivariate logistic regression analyses. Compared with the lowest quartile group of MiBP or MBzP, an incrementally higher risk of depression and moderate/severe depression was observed in the highest quartile (all Ptrend <0.05). It was observed that incrementally higher risk of depression and moderate/severe depression were associated with more numbers of high phthalates parameter (Ptrend <0.001 and Ptrend = 0.003, respectively). A significant interaction between race (Non-Hispanic Black vs. Mexican American) and 2 parameters (having value in the highest quartile of both MiBP and MBzP) was detected for depression (Pinteraction = 0.023) and moderate/severe depression (Pinteraction = 0.029). CONCLUSION: Individuals with more numbers of high phthalates parameter were at higher risk of depression and moderate/severe depression. Non-Hispanic Black participants were more likely to be affected by high levels of MiBP and MBzP exposure than Mexican American participants.

SPRSound: Open-Source SJTU Paediatric Respiratory Sound Database.

It has proved that the auscultation of respiratory sound has advantage in early respiratory diagnosis. Various methods have been raised to perform automatic respiratory sound analysis to reduce subjective diagnosis and physicians' workload. However, these methods highly rely on the quality of respiratory sound database. In this work, we have developed the first open-access paediatric respiratory sound database, SPRSound. The database consists of 2,683 records and 9,089 respiratory sound events from 292 participants. Accurate label is important to achieve a good prediction for adventitious respiratory sound classification problem. A custom-made sound label annotation software (SoundAnn) has been developed to perform sound editing, sound annotation, and quality assurance evaluation. A team of 11 experienced paediatric physicians is involved in the entire process to establish golden standard reference for the dataset. To verify the robustness and accuracy of the classification model, we have investigated the effects of different feature extraction methods and machine learning classifiers on the classification performance of our dataset. As such, we have achieved a score of 75.22%, 61.57%, 56.71%, and 37.84% for the four different classification challenges at the event level and record level.

Using artificial intelligence to reduce queuing time and improve satisfaction in pediatric outpatient service: A randomized clinical trial.

Introduction: Complicated outpatient procedures are associated with excessive paperwork and long waiting times. We aimed to shorten queuing times and improve visiting satisfaction. Methods: We developed an artificial intelligence (AI)-assisted program named Smart-doctor. A randomized controlled trial was conducted at Shanghai Children's Medical Center. Participants were randomly divided into an AI-assisted and conventional group. Smart-doctor was used as a medical assistant in the AI-assisted group. At the end of the visit, an e-medical satisfaction questionnaire was asked to be done. The primary outcome was the queuing time, while secondary outcomes included the consulting time, test time, total time, and satisfaction score. Wilcoxon rank sum test, multiple linear regression and ordinal regression were also used. Results: We enrolled 740 eligible patients (114 withdrew, response rate: 84.59%). The median queuing time was 8.78 (interquartile range [IQR] 3.97,33.88) minutes for the AI-assisted group versus 21.81 (IQR 6.66,73.10) minutes for the conventional group (p < 0.01), and the AI-assisted group had a shorter consulting time (0.35 [IQR 0.18, 0.99] vs. 2.68 [IQR 1.82, 3.80] minutes, p < 0.01), and total time (40.20 [IQR 26.40, 73.80] vs. 110.40 [IQR 68.40, 164.40] minutes, p < 0.01). The overall satisfaction score was increased by 17.53% (p < 0.01) in the AI-assisted group. In addition, multiple linear regression and ordinal regression showed that the queuing time and satisfaction were mainly affected by group (p < 0.01), and missing the turn (p < 0.01). Conclusions: Using AI to simplify the outpatient service procedure can shorten the queuing time of patients and improve visit satisfaction.

The Relationship Between Daily Dietary Intake of Fiber and Short Sleep Duration in the Presence of Di(2-Ethylhexyl) Phthalate: A Population-Based Study.

Objective: This study aimed to evaluate the relationship between daily dietary intake of fiber (DDIF) and short sleep duration (SSD) in the presence of di(2-ethylhexyl) phthalate. Methods: Data of 13,634 participants in this study were collected from the National Health and Nutrition Examination Survey (NHANES). The sum of urinary mono-2-ethyl-5-carboxypentyl phthalate, mono-(2-ethyl-5-hydroxyhexyl) phthalate, mono-(2-ethyl)-hexyl phthalate, and mono-(2-ethyl-5-oxohexyl) phthalate was used to evaluate the level of di(2-ethylhexyl) phthalate (DEHP) exposure. The ln-transformed urinary creatinine-corrected DEHP [ln(DEHP/UCr)] level was used in the statistical models. DDIF was divided into tertiles (<5.77 g/1,000 kcal, 5.77-9.04 g/1,000 kcal, and ≥9.04 g/1,000 kcal). Results: The 13,634 participants included in this study were classified into two groups according to sleep duration. The dose response analysis showed that higher ln(DEHP/UCr) was related to a higher risk of SSD (<7 h and <6 h). Participants in the highest vs. the lowest quartile of DEHP were found to be at increased risk of SSD (<7 h, <6 h, and <5 h). The result of risk of SSD <7 h was OR 1.57, 95% CI, 1.40-1.76; Ptrend <0.001, of SSD <6 h was OR 1.38, 95% CI, 1.18-1.61; Ptrend <0.001, and of SSD <5 h was OR 1.45, 95% CI, 1.13-1.86; Ptrend <0.001. DEHP exposure was found to be associated with SSD <7 h in a sex-specific manner (Pinteraction = 0.033). A significant interaction between ln(DEHP/UCr) and DDIF(tertiles1 vs. tertiles2) (Pinteraction = 0.02) was detected for SSD <7 h. Conclusion: Our results showed that there was a harmful association between DEHP exposure and SSD (<7 h, <6 h, and <5 h). The ameliorative effects of median level of DDIF on SSD <7 h in the presence of DEHP exposure were observed in this study.

Establishment of a Simple Pediatric Lower Respiratory Tract Infections Database Based on the Structured Electronic Medical Records.

Objective: This study aimed to establish a pediatric lower respiratory tract infections (PLRTIs) database based on the structured electronic medical records (SEMRs), to provide a brief overview and the usage process of the SEMRs and the database. Methods: All the medical information is recorded by a clinical information system developed by Eureka Systems Company. A plugin of the software was used to set the properties of items of the SEMR. Children with lower respiratory tract infections (LRTIs) who were admitted to the department of respiratory medicine of our hospital from May 2020 were included. PostgreSQL 13.1 software was used to construct the PLRTIs database. Results: Seven kinds of SEMRs were established, and the admission record was the most important and complex among them. It was mainly composed of 10 parts, i.e., basic information, history of present illness, past history (without respiratory disease), past history of respiratory diseases, personal history, family history, physical examination, the score of LRTIs, auxiliary examination, and diagnosis. With the three-level doctor ward round, the recorded information of the SEMR would be checked repeatedly, thus guaranteeing the correctness of the information. The data of the SEMR and laboratory tests could be extracted directly from the hospital information system (HIS) by PostgreSQL 13.1 software with the specific structured query language (SQL) code. After manually checking the original records, the datasets were imported into PostgreSQL 13.1 software, and a simple PLRTIs database was constructed. According to the inclusion criteria of this study, a total of 1,184 children with LRTIs were included in this database from 1 May 2020 to 30 April 2021. Conclusion: A series of SEMRs for PLRTIs were designed and used during the hospitalization. A simple PLRTIs database was established based on the SEMR. The SEMRs will provide complete and high-quality data on LRTIs in children.

A 2.66 µW Clinician-Like Cardiac Arrhythmia Watchdog Based on P-QRS-T for Wearable Applications.

A wearable electrocardiogram (ECG) device is an effective tool for managing cardiovascular diseases. This paper presents a low power clinician-like cardiac arrhythmia watchdog (CAW) for wearable ECG devices. The CAW is based on a novel P-QRS-T detection algorithm that makes use of clinical features to identify abnormalities. Implemented in 0.18 µm CMOS process, the CAW consumes 2.66 µW for 80 bpm heart rate at 1.2 V supply with an area of 0.578 mm2. Verified on QT database, the average sensitivity/positive predictivity for P-wave, QRS complex and T-wave are over 93.39%/88.55%, 99.69%/99.48%, and 97.13%/93.18% respectively, across over 190000 beats. It shows over 99.8% arrhythmia detection accuracy for 43 subjects evaluated on MIT-BIH database.

The Association between Daily Dietary Intake of Riboflavin and Lung Function Impairment Related with Dibutyl Phthalate Exposure and the Possible Mechanism.

This study aimed to evaluate the relationship between the daily dietary intake of riboflavin (DDIR) and impaired lung function associated with dibutyl phthalate (DBP) exposure. Data of 4631 adults in this national cross-sectional survey were included. Urinary mono-benzyl phthalate (MBP) was used to evaluate the level of DBP exposure. The ln-transformed urinary creatinine-corrected MBP (ln(MBP/UCr)) level was used in the statistical models. High DDIR was defined as the DDIR ≥1.8 mg per day. The results of lung function impairment and high monocytes were significantly higher in the highest MBP group compared with the lowest MBP group. A significant interaction between ln(MBP/UCr) and DDIR (Pinteraction = 0.029) was detected for the risk of lung function impairment. The risk of lung function impairment (ORquartiles4 vs. 1 1.85, 95% CI, 1.27-2.71; Ptrend = 0.018) and high neutrophils (ORquartiles4 vs. 1 1.45, 95% CI, 1.06-1.97; Ptrend = 0.018) was significantly higher in the highest vs. the lowest quartile of MBP in participants with low/normal DDIR but not in in participants with high DDIR. The results of this study showed that high DDIR was associated with less lung function impairment related with DBP exposure, and the inhibiting of the neutrophil recruitment might be the potential mechanism.

Automatic Detection of Secundum Atrial Septal Defect in Children Based on Color Doppler Echocardiographic Images Using Convolutional Neural Networks.

Secundum atrial septal defect (ASD) is one of the most common congenital heart diseases (CHDs). This study aims to evaluate the feasibility and accuracy of automatic detection of ASD in children based on color Doppler echocardiographic images using convolutional neural networks. In this study, we propose a fully automatic detection system for ASD, which includes three stages. The first stage is used to identify four target echocardiographic views (that is, the subcostal view focusing on the atrium septum, the apical four-chamber view, the low parasternal four-chamber view, and the parasternal short-axis view). These four echocardiographic views are most useful for the diagnosis of ASD clinically. The second stage aims to segment the target cardiac structure and detect candidates for ASD. The third stage is to infer the final detection by utilizing the segmentation and detection results of the second stage. The proposed ASD detection system was developed and validated using a training set of 4,031 cases containing 370,057 echocardiographic images and an independent test set of 229 cases containing 203,619 images, of which 105 cases with ASD and 124 cases with intact atrial septum. Experimental results showed that the proposed ASD detection system achieved accuracy, recall, precision, specificity, and F1 score of 0.8833, 0.8545, 0.8577, 0.9136, and 0.8546, respectively on the image-level averages of the four most clinically useful echocardiographic views. The proposed system can automatically and accurately identify ASD, laying a good foundation for the subsequent artificial intelligence diagnosis of CHDs.

Quality and efficiency of a standardized e-handover system for pediatric nursing: A prospective interventional study.

AIM: This study examined the effect on pediatric nursing handover quality and efficiency when a standardized e-handover system was implemented. BACKGROUND: Handover quality is an important aspect of nursing quality management; however, handover quality among nursing staff is poor. METHODS: A prospective interventional study was carried out in a general pediatrics ward from December 2019 to November 2020. The tools included a standardized e-handover system. The intervention strategies included workflow remodeling and employee training on oral handover using the standardized e-handover system. RESULTS: The omission frequency of critical handover elements decreased from 47.32% to 2.94% (p < .01), among which the omission frequencies of nine out of 16 key elements significantly decreased. Integrity also showed improvement. Specifically, the integrity of five types of critical information was significantly improved, including vital signs, signs and symptoms, laboratory test results, radiologic examination results, and treatment regimen (2.00 vs. 5.00, p < .01; 3.00 vs. 5.00, p < .01; 3.00 vs. 5.00, p < .01; 5.00 vs. 5.00, p = .009; 3.00 vs. 4.00, p < .01, respectively). Information accuracy was 100%. Workflow and efficiency significantly improved, communication duration with patient/family during work hours significantly increased (24.00 vs. 56.00, p < .01), and prehandover preparation duration significantly decreased (32.00 vs. 2.50, p < .01). Nurse handover satisfaction showed improvement (56.88 ± 15.08 vs. 74.31 ± 9.22, p < .01). CONCLUSION: The standardized e-handover system effectively improved nurse handover quality, optimized workflow, increased work efficiency, and promoted teamwork. IMPLICATIONS FOR NURSING MANAGEMENT: Standardized e-handover systems have great potential for ensuring the safety of pediatric patients and improving the quality of handover.

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.

Pre-Consultation System Based on the Artificial Intelligence Has a Better Diagnostic Performance Than the Physicians in the Outpatient Department of Pediatrics.

Artificial intelligence (AI) has been deeply applied in the medical field and has shown broad application prospects. Pre-consultation system is an important supplement to the traditional face-to-face consultation. The combination of the AI and the pre-consultation system can help to raise the efficiency of the clinical work. However, it is still challenging for the AI to analyze and process the complicated electronic health record (EHR) data. Our pre-consultation system uses an automated natural language processing (NLP) system to communicate with the patients through the mobile terminals, applying the deep learning (DL) techniques to extract the symptomatic information, and finally outputs the structured electronic medical records. From November 2019 to May 2020, a total of 2,648 pediatric patients used our model to provide their medical history and get the primary diagnosis before visiting the physicians in the outpatient department of the Shanghai Children's Medical Center. Our task is to evaluate the ability of the AI and doctors to obtain the primary diagnosis and to analyze the effect of the consistency between the medical history described by our model and the physicians on the diagnostic performance. The results showed that if we do not consider whether the medical history recorded by the AI and doctors was consistent or not, our model performed worse compared to the physicians and had a lower average F1 score (0.825 vs. 0.912). However, when the chief complaint or the history of present illness described by the AI and doctors was consistent, our model had a higher average F1 score and was closer to the doctors. Finally, when the AI had the same diagnostic conditions with doctors, our model achieved a higher average F1 score (0.931) compared to the physicians (0.92). This study demonstrated that our model could obtain a more structured medical history and had a good diagnostic logic, which would help to improve the diagnostic accuracy of the outpatient doctors and reduce the misdiagnosis and missed diagnosis. But, our model still needs a good deal of training to obtain more accurate symptomatic information.

Mycoplasma pneumoniae infection prediction model for hospitalized community-acquired pneumonia children.

OBJECTIVES: We sought to develop a nomogram to predict Mycoplasma pneumoniae (Mp) infection among hospitalized children with community-acquired pneumonia (CAP) and compare it with another model developed from age and duration of fever. METHODS: Data on 5904 CAP children who were enrolled at Shanghai Children's Medical Center were retrospectively collected and divided into a training set (n = 4133) and a validation set (n = 1771). The model's performance was determined by concordance index (C-index), calibration curves, Brier scores, and decision curve analyses (DCAs). Akaike information criterion (AIC) and Bayesian information criterion (BIC) were used for model comparisons. RESULTS: Incorporating five factors (age, duration of fever, erythrocyte sedimentation rate, leukocyte count, and neutrophil proportion), the nomogram achieved good C-index values of 0.74 (95% confidence interval [CI]: 0.72-0.76) and 0.75 (95% CI: 0.73-0.78) and good Brier scores of 0.14 (95% CI: 0.13-0.15) and 0.17 (95% CI: 0.15-0.18) in predicting Mp infection in the training and validation cohorts, respectively, and had moderate fitted calibration plots. The DCAs showed good clinical usefulness of the nomogram. Patients were effectively divided into low, medium, and high risk groups by two cut-off score points of the nomogram, 210 and 300. With the lower AIC (3673.5) and BIC (3774.7) value, the model of five predictors is the better model. CONCLUSIONS: By using five predictor variables, a simple nomogram of good predictive accuracy for Mp infection and moderate agreements between the actual outcome and the predicted probability was constructed. It could serve as a tool to aid physicians in clinical decision-making processes.

LungAttn: advanced lung sound classification using attention mechanism with dual TQWT and triple STFT spectrogram.

Objective. Auscultation of lung sound plays an important role in the early diagnosis of lung diseases. This work aims to develop an automated adventitious lung sound detection method to reduce the workload of physicians.Approach. We propose a deep learning architecture, LungAttn, which incorporates augmented attention convolution into ResNet block to improve the classification accuracy of lung sound. We adopt a feature extraction method based on dual tunableQ-factor wavelet transform and triple short-time Fourier transform to obtain a multi-channel spectrogram. Mixup method is introduced to augment adventitious lung sound recordings to address the imbalance dataset problem.Main results. Based on the ICBHI 2017 challenge dataset, we implement our framework and compare with the state-of-the-art works. Experimental results show that LungAttn has achieved theSensitivity, Se,Specificity, SpandScoreof 36.36%, 71.44% and 53.90%, respectively. Of which, our work has improved theScoreby 1.69% compared to the state-of-the-art models based on the official ICBHI 2017 dataset splitting method.Significance. Multi-channel spectrogram based on different oscillatory behavior of adventitious lung sound provides necessary information of lung sound recordings. Attention mechanism is introduced to lung sound classification methods and has proved to be effective. The proposed LungAttn model can potentially improve the speed and accuracy of lung sound classification in clinical practice.

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.

Artificial Intelligence for Screening Chinese Electronic Medical Record and Biobank Information.

Objective: To establish a structured and integrated platform of clinical data and biobank data, and a client to retrieve these data. Study Design: Initially, the hospital information system (HIS) and biobank information system (BIS) were integrated through the patients' ID numbers. Then, natural language processing (NLP) was used to process the integrated unstructured clinical information. A query interface was designed for this system, which enabled researchers to retrieve clinical or biobank data. Finally, several queries were listed and manually checked to test the retrieval performance of the system. Results: The construction of the biobank screening system (BSS) was completed, and the data were structured. The BSS took an average of 2 seconds to perform a search for target patients/samples. The retrieval results were consistent with the HIS and BIS. For complex queries, we manually checked the retrieved patients/samples, and the system's accuracy was 100%. Conclusion: This NLP-based system improved biological sample screening and using of clinical data. We will continue to improve this system, enhance resource sharing, and promote the development of translational medicine.

Real-World Verification of Artificial Intelligence Algorithm-Assisted Auscultation of Breath Sounds in Children.

Objective: Lung auscultation plays an important role in the diagnosis of pulmonary diseases in children. The objective of this study was to evaluate the use of an artificial intelligence (AI) algorithm for the detection of breath sounds in a real clinical environment among children with pulmonary diseases. Method: The auscultations of breath sounds were collected in the respiratory department of Shanghai Children's Medical Center (SCMC) by using an electronic stethoscope. The discrimination results for all chest locations with respect to a gold standard (GS) established by 2 experienced pediatric pulmonologists from SCMC and 6 general pediatricians were recorded. The accuracy, sensitivity, specificity, precision, and F1-score of the AI algorithm and general pediatricians with respect to the GS were evaluated. Meanwhile, the performance of the AI algorithm for different patient ages and recording locations was evaluated. Result: A total of 112 hospitalized children with pulmonary diseases were recruited for the study from May to December 2019. A total of 672 breath sounds were collected, and 627 (93.3%) breath sounds, including 159 crackles (23.1%), 264 wheeze (38.4%), and 264 normal breath sounds (38.4%), were fully analyzed by the AI algorithm. The accuracy of the detection of adventitious breath sounds by the AI algorithm and general pediatricians with respect to the GS were 77.7% and 59.9% (p < 0.001), respectively. The sensitivity, specificity, and F1-score in the detection of crackles and wheeze from the AI algorithm were higher than those from the general pediatricians (crackles 81.1 vs. 47.8%, 94.1 vs. 77.1%, and 80.9 vs. 42.74%, respectively; wheeze 86.4 vs. 82.2%, 83.0 vs. 72.1%, and 80.9 vs. 72.5%, respectively; p < 0.001). Performance varied according to the age of the patient, with patients younger than 12 months yielding the highest accuracy (81.3%, p < 0.001) among the age groups. Conclusion: In a real clinical environment, children's breath sounds were collected and transmitted remotely by an electronic stethoscope; these breath sounds could be recognized by both pediatricians and an AI algorithm. The ability of the AI algorithm to analyze adventitious breath sounds was better than that of the general pediatricians.

Is there possibility of vertical transmission of COVID-19: a systematic review.

In order to investigate the clinical features of pregnant women and their neonates with coronavirus disease 2019 (COVID-19) and the evidence of vertical transmission of COVID-19, we retrieved studies included in PubMed, Medline and Chinese databases from January 1, 2000 to October 25, 2020 using relevant terms, such as 'COVID-19', 'vertical transmission' et al. in 'Title/Abstract'. Case reports and case series were included according to the inclusion and exclusion criteria. We conducted literature screening and data extraction, and performed literature bias risk assessment. Total of 13 case series and 16 case reports were collected, including a total of 564 pregnant women with COVID-19 and their 555 neonates, of which 549 neonates received nucleic acid test for the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and 18 neonates was diagnosed with COVID-19. The positive rate is 3.28%. Amniotic fluid of one woman was tested positive for SARS-CoV-2. The majority of infected neonates were born under strict infection control and received isolation and artificial feeding. Up till now, there is no sufficient evidence to exclude the possibility of vertical transmission for COVID-19 based on the current available data.