Quantitative muscle ultrasound in children with Duchenne muscular dystrophy: Comparing to magnetic resonance imaging.

OBJECTIVE: There is a need today for favorable biomarkers to follow up on the disease progression and therapeutic response in patients with Duchenne muscular dystrophy (DMD). This study evaluates whether quantitative muscle ultrasound (QMUS) or magnetic resonance imaging (MRI) is more suitable for the assessment of DMD in China. METHODS: Thirty-six boys with DMD, who were treated with prednisone from baseline to month 12, were enrolled in this longitudinal, observational cohort study. Muscle thickness and echo intensity on QMUS and T1-weighted MRI grading were measured in the right rectus femoris. RESULTS: Scores for muscle thickness and echo intensity in QMUS and T1-weighted MRI grading showed significant correlations with the clinical characteristics of muscle strength, timed testing, and quality of life (p < 0.05). Scores for muscle thickness and echo intensity on QMUS also showed good correlations with T1-weighted MRI grading (p < 0.05). However, 15 of 36 boys with DMD did not undergo MRI examinations for various reasons. CONCLUSIONS: QMUS and MRI can be used as biomarkers for tracking DMD to some extent. Both have strengths and weaknesses and the specific needs and goals of the clinical or research project are what make one preferable to the other.

Be aware of acute kidney injury in critically ill children with COVID-19.

BACKGROUND: Acute kidney injury (AKI) is a common complication of critically ill adult patients with COVID-19. However, currently, no studies investigate kidney impairment in children with COVID-19. We investigated incidence and treatment of AKI in pediatric patients with COVID-19 in Wuhan Children's Hospital during the early stages of the COVID-19 pandemic and discuss possible mechanisms of AKI related to SARS-CoV-2 infection. METHODS: By extracting data from electronic medical records, we conducted a retrospective observational study of kidney involvement in confirmed pediatric COVID-19 cases in Wuhan Children's Hospital during the coronavirus outbreak, from January 24 to March 20, 2020. Clinical presentations, clinical courses, laboratory findings, and medical interventions are described below. RESULTS: Among 238 confirmed COVID-19 cases, only three were critically ill and needed intensive care unit (ICU) admission. All three developed AKI, but AKI was not detected in any non-critically ill patients outside the ICU. Two of the three patients with AKI had prodromal gastrointestinal symptoms. Significantly elevated interleukin-6 (IL-6) levels and complement activation were observed in these patients with AKI. The three patients with AKI were treated with plasma exchange (PE) and continuous kidney replacement therapy (CKRT), resulting in one complete recovery, one partial recovery, and one mortality due to critical illness. CONCLUSIONS: Critically ill children with COVID-19 may develop AKI, especially following prodromal gastrointestinal symptoms. An inflammatory storm and complement-mediated injury may underlie AKI development in children with COVID-19. Our study supports implantation of PE and CKRT in management of critically ill patients with AKI.

Overview of chest involvement at computed tomography in children with coronavirus disease 2019 (COVID-19).

BACKGROUND: Chest computed tomography (CT) findings in children with coronavirus disease 2019 (COVID-19) have been rarely reported in a comprehensive and systematic manner. OBJECTIVE: We investigated the chest CT findings in children with COVID-19, and explored the differences in these findings between symptomatic patients and asymptomatic patients. MATERIALS AND METHODS: Demographic findings, clinical characteristics, duration of hospital stay and viral shedding, and chest CT findings in 201 children infected with severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) were retrospectively analyzed from January 15 to March 20, 2020, and divided into two groups: symptomatic group (n=136) and asymptomatic group (n=65). Chi-square test and Student's t-test were used for statistical analysis. RESULTS: Symptomatic patients were mainly young children ≤3 years old (54/63, 86%),while asymptomatic patients were mainly children ≥ 6 years old (51/111, 46%). Fever (41%) and cough (41%) were the most common symptoms. Overall, 119/201 (59%) patients had chest CT findings, and symptomatic patients accounted for 82% (98/119). The CT findings presented as bilateral multiple lesions (60/119, 50.4%), ground-glass opacities (83/119, 70%) and/or consolidation (44/119, 37%) with a peripheral and subpleural distribution (62/83, 75%). Fifteen of 87 (7.2%) patients with lung lesions showed complete lesion absorption, and 42/87 (48%) improved within a mean of 9.1 (standard deviation [SD] 3.2) days. The mean duration of viral shedding was 8.7 (SD 4.9) days. Pleural effusion was very rare. No lymphadenopathy was found in either group. CONCLUSION: Symptoms associated with pulmonary involvement were most common in infants and young children. The lung lesions of most patients were absorbed and improved in about 9 days.

Identification and characterization of novel mutations in MOGS in a Chinese patient with infantile spams.

Congenital disorders of glycosylation (CDGs) are a genetically heterogeneous group of disorders caused by the defects in the synthesis and processing of glycoproteins. CDG is caused by mannosyl-oligosaccharide glucosidase (MOGS) deficiency, and is an extremely rare type, and only six patients have been reported. Here, we report a patient from China with facial dysmorphism, infantile spams, developmental delay, low vison, and abnormal liver function and low immunoglobulin. Brain MRI showed hypoplasia of the corpus callosum and slightly wide sulci at bilateral frontal parietal lobes. Compound heterozygous mutations of (c.1694G>A: R565Q and c.1619G>A: R540H) in exon 4 of MOGS gene (NM_006302.2) were identified by whole exome sequencing. Further investigation showed that the gene expression of MOGS in patients' peripheral blood was decreased. We observed that two mutations were associated with lower protein expression of MOGS, cell growth, and cell cycle in transiently transfected Hela cells. We also noticed that cell cycle-related proteins, ß-catenin, cyclin D1, and C-myc, were decreased in mutant cells. In conclusion, our study suggested whole exome sequencing, and genes associated with CDGs should be analyzed in patients with infantile spams and multiple system involvement, and mutant MOGS-impaired cell cycle progression. Our work broadens the mutation spectrum of MOGS gene.

Association of Web-Based Physical Education With Mental Health of College Students in Wuhan During the COVID-19 Outbreak: Cross-Sectional Survey Study.

BACKGROUND: The COVID-19 outbreak has affected people's health worldwide. For college students, web-based physical education is a challenge, as these course are normally offered outdoors. OBJECTIVE: The aim of this study was to use data from a web-based survey to evaluate the relationship between the mental health status of college students and their sports-related lifestyles. Problems related to web-based physical education were also examined. METHODS: A web-based survey was conducted by snowball sampling from May 8 to 11, 2020. Demographic data, mental health status, and sports-related lifestyles of college students in Wuhan as well as issues related to web-based physical education were collected. Mental health status was assessed by the Depression, Anxiety, and Stress Scale (DASS-21). RESULTS: The study included 1607 respondents from 267 cities. The average scores of the DASS-21 subscales (2.46 for depression, 1.48 for anxiety, and 2.59 for stress) were significantly lower in our study than in a previous study (P<.05). Lower DASS-21 scores were significantly correlated with regular exercise, maintaining exercise habits during the outbreak of COVID-19, exercising more than 1 to 2 times a week, exercise duration >1 hour, and >2000 pedometer steps (all P<.05). None of the three forms of web-based physical education was preferred by more than 50% of respondents. Frequent technical problems were confronted by 1087/1607 students (67.6%). Shape-up exercises (846/1607, 52.6%), a designed combination of exercises (710/1607, 44.2%), and Chinese kung fu (559/1607, 34.8%) were suggested sports for web-based physical education. CONCLUSIONS: Mental status was significantly correlated with regular exercise and sufficient exercise duration. Professional physical guidance is needed for college students in selected sports. Exercises not meeting students' preferences, frequent technical problems, and the distant interaction involved in web-based physical education were the main problems that should be solved in future.

Distribution and expression of Pen-2 in the central nervous system of APP/PS1 double transgenic mice.

The γ-secretase complex catalyzes the final cleavage step of amyloid ß-protein precursor (APP) to generate amyloid ß (Aß) peptide, a pathogenic component of senile plaques in the brain of Alzheimer's disease (AD) patients. Recent studies have shown that presenilin enhancer-2 (Pen-2), presenilin (PS, including PS1 and PS2), nicastrin, and anterior pharynx-defective 1 are essential components of the γ-secretase. The structure and function of Pen-2 in vitro have been well defined. However, little is known about the neuroanatomical distribution and expression of Pen-2 in the central nervous system (CNS) of AD model mice. We report here, using various methods such as immunohistochemical staining and immunoblotting, that Pen-2 is widely expressed at specific neuronal cells of major areas in AD model mice, including the olfactory bulb, basal forebrain, striatum, cortex, hippocampus, amygdala, thalamus, hypothalamus, cerebellum, brainstem, and spinal cord. It is co-expressed with PS1 in specific neuronal cells in mouse brain. Pen-2 is distributed much more extensively than extracellular amyloid deposits, suggesting the importance of other factors in localized amyloid deposition. Pen-2 is localized predominantly in cell membrane and cytoplasma in adult AD mice, but only distributed at cell membrane in controls. At the early stages of postnatal development, the expression level of Pen-2 is relatively high in CNS, but declines, gradually in adult mice. The present study provides an anatomical basis for Pen-2 as a key component of γ-secretase complex in the brain of developing and adult mice, and Pen-2 might be closely related to Aß burden in aging nervous system.

Multimodal imaging of congenital pyriform fossa fistula in children.

Objective: Our aim was to explore the clinical value of multimodal imaging examinations in the diagnosis of congenital pyriform fossa fistula in children, so as to provide clues for the early diagnosis and treatment of congenital pyriform fossa fistula. Methods: The clinical and imaging data of 55 children with pyriform fossa fistula diagnosed surgically in our hospital from 2015 to 2018 were analyzed retrospectively. All 55 patients underwent a CT scan. Of those patients, contrast enhancement CT was performed in 47 cases, MRI was performed in 2 cases, and barium esophagography was performed in 41 cases. Results: Among the 55 cases, there were 24 male patients and 31 female patients. The age ranged from 11 months to 13 years old, and the median age was 3.8 years old. The lesions of 49 cases (89.1%) were located on the left side, and the imaging of CT showed soft tissue mass in the anterior cervical region blurred boundary. There was ipsilateral thyroid involvement in 50 cases (90.9%), trachea and/or carotid sheath extension in 43 cases (78.2%), abscess formation in 39 cases (70.9%), and pneumatosis in 25 cases (45.5%). The CT examination of 22 children after treatment showed a linear or tubular low-density shadow in the thyroid gland, gas accumulation in the anterior cervical region or thyroid, and residual contrast medium, partly. A total of 24 cases underwent barium esophagography during the acute phase, and 15 cases (62.5%) showed sinus formation from the pyriform fossa downward or punctate high-density shadow in the anterior cervical region. The 2 cases where MRI was performed showed abscess formation in one side of the neck and thyroid involvement. Conclusion: Pyriform fossa fistula is most common in the left anterior cervical region, and it is closely related to the thyroid gland. The plain and enhanced-contrast CT scan can be used as the first choice during the infection stage. It helps to understand the location, extent, and structure of the surrounding tissue. The preliminary diagnosis of pyriform sinus fistula was according to the imaging features. It provided an important basis for clinical diagnosis and reduced the pain caused by repeated infection or surgical incision and drainage.

Almost misdiagnosed Menkes disease: A case report.

Background: Menkes disease is a disorder of copper metabolism and which follows a progressive degeneration of brain. It is a rare X-linked recessive disorder that results from mutations in ATP7A gene. The early diagnosis of Menkes disease is critical to patients' prognosis. Case presentation: We report a case of Menkes disease. A 4-month-old boy presented with intermittent convulsions for a week. The brain MRI showed excessive tortuosities of intracranial vessels, and radiologists prompted for further examinations to confirm that it was Menkes disease. Patient was advised for biochemical investigations and genetic tests. Reduced level of ceruloplasmin (0.04 g/L; normal range, 0.2-0.6 g/L) was revealed. Genetic testing revealed a missense mutation within exon 18, c.3548 G > A, p.G1183D. This patient was almost misdiagnosed as epilepsy. Fortunately, based on the clues from radiologist, further physical examination and experimental tests were carried out. Conclusion: We reported the imaging features of a case of Menkes disease, which can provide clinicians with more clues to consider the possibility of this rare disease.

Erratum: Author Correction: Advancing COVID-19 diagnosis with privacy-preserving collaboration in artificial intelligence.

[This corrects the article DOI: 10.1038/s42256-021-00421-z.].

Allergic Asthma-Induced Cognitive Impairment is Alleviated by Dexamethasone.

Allergic asthma is a typical chronic inflammatory disease of respiratory tract. Clinical data shows that patients with allergic asthma have different degrees of cognitive dysfunction. The molecular mechanism underlying the pathogenesis of asthma-induced cognitive disorder is not yet well defined. Dexamethasone (DEX), one of the first-line drugs being widely used in the treatment of asthma, has not been reported to have an effect on cognitive dysfunction in mice model. To investigate the effect of asthma on cognitive impairment as well as the effect of DEX on asthma-caused morphological and behavioral changes, C57BL/6J mice received treatment with house dust mites (HDM) for 60 days to become allergic asthma model mice, and a group of HDM-treated asthma model mice were treated with DEX. HDM-treated asthma model mice exhibited increased airway hyperresponsiveness (AHR) and inflammatory infiltration in lung tissue. An elevated level of IL-4, IL-5, and TNF-α was detected in bronchoalveolar lavage fluid (BALF) by Luminex liquid suspension chip. Asthma model mice also presented memory deficits accompanied with morphological changes at the synaptic levels in the cortex and hippocampus. Meanwhile, vascular edema and increased expression of HIF-1α and HIF-2α were found in the brain of asthma model mice. Interestingly, DEX treatment could reverse the inflammatory changes in asthma model mice airway, rescue the cognitive impairment and improve the synaptic plasticity. Besides, DEX significantly decreased the expression of HIF-1α and HIF-2α in mice brain and lung. These processes may be used to decipher the complex interplay and pathological changes between asthma and cognition. This study provides laboratory evidence for the prevention and treatment of cognitive malfunction induced by asthma.

Quantitative assessment of COVID-19 pneumonia in neonates using lung ultrasound score.

BACKGROUND: Lung ultrasound (LUS) and lung ultrasound score (LUSS) have been successfully used to diagnose neonatal pneumonia, assess the lesion distribution, and quantify the aeration loss. The present study design determines the diagnostic value of LUSS in the semi-quantitative assessment of pneumonia in coronavirus disease 2019 (COVID-19) neonates. METHODS: Eleven COVID-19 neonates born to mothers with COVID-19 infection and 11 age- and gender-matched controls were retrospectively studied. LUSS was acquired by assessing the lesions and aeration loss in 12 lung regions per subject. RESULTS: Most of the COVID-19 newborns presented with mild and atypical symptoms, mainly involving respiratory and digestive systems. In the COVID-19 group, a total of 132 regions of the lung were examined, 83 regions (62.8%) of which were detected abnormalities by LUS. Compared with controls, COVID-19 neonates showed sparse or confluent B-lines (83 regions), disappearing A-lines (83 regions), abnormal pleural lines (29 regions), and subpleural consolidations (2 regions). The LUSS was significantly higher in the COVID-19 group. In total, 49 regions (37%) were normal, 73 regions (55%) scored 1, and 10 regions (8%) scored 2 by LUSS. All the lesions were bilateral, with multiple regions involved. The majority of the lesions were located in the bilateral inferior and posterior regions. LUS detected abnormalities in three COVID-19 neonates with normal radiological performance. The intra-observer and inter-observer reproducibility of LUSS was excellent. CONCLUSIONS: LUS is a noninvasive, convenient, and sensitive method to assess neonatal COVID-19 pneumonia, and can be used as an alternative to the use of diagnostic radiography. LUSS provides valuable semi-quantitative information on the lesion distribution and severity.

Advancing COVID-19 Diagnosis with Privacy-Preserving Collaboration in Artificial Intelligence.

Artificial intelligence (AI) provides a promising substitution for streamlining COVID-19 diagnoses. However, concerns surrounding security and trustworthiness impede the collection of large-scale representative medical data, posing a considerable challenge for training a well-generalised model in clinical practices. To address this, we launch the Unified CT-COVID AI Diagnostic Initiative (UCADI), where the AI model can be distributedly trained and independently executed at each host institution under a federated learning framework (FL) without data sharing. Here we show that our FL model outperformed all the local models by a large yield (test sensitivity /specificity in China: 0.973/0.951, in the UK: 0.730/0.942), achieving comparable performance with a panel of professional radiologists. We further evaluated the model on the hold-out (collected from another two hospitals leaving out the FL) and heterogeneous (acquired with contrast materials) data, provided visual explanations for decisions made by the model, and analysed the trade-offs between the model performance and the communication costs in the federated training process. Our study is based on 9,573 chest computed tomography scans (CTs) from 3,336 patients collected from 23 hospitals located in China and the UK. Collectively, our work advanced the prospects of utilising federated learning for privacy-preserving AI in digital health.

Advancing COVID-19 diagnosis with privacy-preserving collaboration in artificial intelligence.

Artificial intelligence provides a promising solution for streamlining COVID-19 diagnoses; however, concerns surrounding security and trustworthiness impede the collection of large-scale representative medical data, posing a considerable challenge for training a well-generalized model in clinical practices. To address this, we launch the Unified CT-COVID AI Diagnostic Initiative (UCADI), where the artificial intelligence (AI) model can be distributedly trained and independently executed at each host institution under a federated learning framework without data sharing. Here we show that our federated learning framework model considerably outperformed all of the local models (with a test sensitivity/specificity of 0.973/0.951 in China and 0.730/0.942 in the United Kingdom), achieving comparable performance with a panel of professional radiologists. We further evaluated the model on the hold-out (collected from another two hospitals without the federated learning framework) and heterogeneous (acquired with contrast materials) data, provided visual explanations for decisions made by the model, and analysed the trade-offs between the model performance and the communication costs in the federated training process. Our study is based on 9,573 chest computed tomography scans from 3,336 patients collected from 23 hospitals located in China and the United Kingdom. Collectively, our work advanced the prospects of utilizing federated learning for privacy-preserving AI in digital health.

Features Discriminating COVID-19 From Community-Acquired Pneumonia in Pediatric Patients.

Purpose: To discuss the different characteristics of clinical, laboratory and chest computed tomography (CT) between coronavirus disease 2019 (COVID-19) and community-acquired pneumonia (CAP) in pediatric patients. Methods: We retrospectively retrieved data of inpatients with COVID-19 from January 21st to March 14th, 2020, and CAP from November 1st, 2019 to December 31st, 2019 in Wuhan Children's Hospital. We divided CAP into mycoplasma pneumonia and other viral pneumonia. We analyzed clinical and radiological features from those patients, and compared the differences among COVID-19, mycoplasma pneumonia and other viral pneumonia. Results: Eighty COVID-19 inpatients from January 21st to March 14th, 2020, as well as 95 inpatients with mycoplasma pneumonia and 50 inpatients with other viral pneumonia from November 1st, 2019 to December 31st, 2019 were included in our study. All patients were confirmed with RT-PCR. The clinical symptoms were similar in the three groups. Except fever and cough, diarrhea (6/80, 7.5%), tachypnea (2/80, 2.5%), and fatigue (6/80, 7.5%) were less common in COVID-19 patients. Compared to mycoplasma pneumonia and other viral pneumonia inpatients, COVID-19 patients present remarkably increased alanine aminotransferase (69/80, 86.3%). The typical CT feature of COVID-19 is ground-glass opacity, and it was more common in COVID-19 patients (32/80, 40%). Conclusion: The COVID-19 shared similar onsets with CAP. Even though the ground-glass opacity and elevated level of ALT were frequent in COVID-19, the better way for treatment and management of this disease is quickly and accurately identifying the pathogen.

Clinical and CT features in pediatric patients with COVID-19 infection: Different points from adults.

PURPOSE: To discuss the different characteristics of clinical, laboratory, and chest computed tomography (CT) in pediatric patients from adults with 2019 novel coronavirus (COVID-19) infection. METHODS: The clinical, laboratory, and chest CT features of 20 pediatric inpatients with COVID-19 infection confirmed by pharyngeal swab COVID-19 nucleic acid test were retrospectively analyzed during 23 January and 8 February 2020. The clinical and laboratory information was obtained from inpatient records. All the patients were undergone chest CT in our hospital. RESULTS: Thirteen pediatric patients (13/20, 65%) had an identified history of close contact with COVID-19 diagnosed family members. Fever (12/20, 60%) and cough (13/20, 65%) were the most common symptoms. For laboratory findings, procalcitonin elevation (16/20, 80%) should be pay attention to, which is not common in adults. Coinfection (8/20, 40%) is common in pediatric patients. A total of 6 patients presented with unilateral pulmonary lesions (6/20, 30%), 10 with bilateral pulmonary lesions (10/20, 50%), and 4 cases showed no abnormality on chest CT (4/20, 20%). Consolidation with surrounding halo sign was observed in 10 patients (10/20, 50%), ground-glass opacities were observed in 12 patients (12/20, 60%), fine mesh shadow was observed in 4 patients (4/20, 20%), and tiny nodules were observed in 3 patients (3/20, 15%). CONCLUSION: Procalcitonin elevation and consolidation with surrounding halo signs were common in pediatric patients which were different from adults. It is suggested that underlying coinfection may be more common in pediatrics, and the consolidation with surrounding halo sign which is considered as a typical sign in pediatric patients.

Dihydroartemisinin Ameliorates Learning and Memory in Alzheimer's Disease Through Promoting Autophagosome-Lysosome Fusion and Autolysosomal Degradation for Aß Clearance.

Dihydroartemisinin (DHA) is an active metabolite of sesquiterpene trioxane lactone extracted from Artemisia annua, which is used to treat malaria worldwide. DHA can activate autophagy, which is the main mechanism to remove the damaged cell components and recover the harmful or useless substances from eukaryotic cells and maintain cell viability through the autophagy lysosomal degradation system. Autophagy activation and autophagy flux correction are playing an important neuroprotective role in the central nervous system, as they accelerate the removal of toxic protein aggregates intracellularly and extracellularly to prevent neurodegenerative processes, such as Alzheimer's disease (AD). In this study, we explored whether this mechanism can mediate the neuroprotective effect of DHA on the AD model in vitro and in vivo. Three months of DHA treatment improved the memory and cognitive impairment, reduced the deposition of amyloid ß plaque, reduced the levels of Aß40 and Aß42, and ameliorated excessive neuron apoptosis in APP/PS1 mice brain. In addition, DHA treatment increased the level of LC3 II/I and decreased the expression of p62. After Bafilomycin A1 and Chloroquine (CQ) blocked the fusion of autophagy and lysosome, as well as the degradation of autolysosomes (ALs), DHA treatment increased the level of LC3 II/I and decreased the expression of p62. These results suggest that DHA treatment can correct autophagic flux, improve autophagy dysfunction, inhibit abnormal death of neurons, promote the clearance of amyloid-ß peptide (Aß) fibrils, and have a multi-target effect on the neuropathological process, memory and cognitive deficits of AD.

Frequency-Specific Changes of Resting Brain Activity in Parkinson's Disease: A Machine Learning Approach.

The application of resting state functional MRI (RS-fMRI) in Parkinson's disease (PD) was widely performed using standard statistical tests, however, the machine learning (ML) approach has not yet been investigated in PD using RS-fMRI. In current study, we utilized the mean regional amplitude values as the features in patients with PD (n = 72) and in healthy controls (HC, n = 89). The t-test and linear support vector machine were employed to select the features and make prediction, respectively. Three frequency bins (Slow-5: 0.0107-0.0286 Hz; Slow-4: 0.0286-0.0821 Hz; conventional: 0.01-0.08 Hz) were analyzed. Our results showed that the Slow-4 may provide important information than Slow-5 in PD, and it had almost identical classification performance compared with the Combined (Slow-5 and Slow-4) and conventional frequency bands. Similar with previous neuroimaging studies in PD, the discriminative regions were mainly included the disrupted motor system, aberrant visual cortex, dysfunction of paralimbic/limbic and basal ganglia networks. The lateral parietal lobe, such as right inferior parietal lobe (IPL) and supramarginal gyrus (SMG), was detected as the discriminative features exclusively in Slow-4. Our findings, at the first time, indicated that the ML approach is a promising choice for detecting abnormal regions in PD, and a multi-frequency scheme would provide us more specific information.

A collaborative online AI engine for CT-based COVID-19 diagnosis.

Artificial intelligence can potentially provide a substantial role in streamlining chest computed tomography (CT) diagnosis of COVID-19 patients. However, several critical hurdles have impeded the development of robust AI model, which include deficiency, isolation, and heterogeneity of CT data generated from diverse institutions. These bring about lack of generalization of AI model and therefore prevent it from applications in clinical practices. To overcome this, we proposed a federated learning-based Unified CT-COVID AI Diagnostic Initiative (UCADI, http://www.ai-ct-covid.team/), a decentralized architecture where the AI model is distributed to and executed at each host institution with the data sources or client ends for training and inferencing without sharing individual patient data. Specifically, we firstly developed an initial AI CT model based on data collected from three Tongji hospitals in Wuhan. After model evaluation, we found that the initial model can identify COVID from Tongji CT test data at near radiologist-level (97.5% sensitivity) but performed worse when it was tested on COVID cases from Wuhan Union Hospital (72% sensitivity), indicating a lack of model generalization. Next, we used the publicly available UCADI framework to build a federated model which integrated COVID CT cases from the Tongji hospitals and Wuhan Union hospital (WU) without transferring the WU data. The federated model not only performed similarly on Tongji test data but improved the detection sensitivity (98%) on WU test cases. The UCADI framework will allow participants worldwide to use and contribute to the model, to deliver a real-world, globally built and validated clinic CT-COVID AI tool. This effort directly supports the United Nations Sustainable Development Goals' number 3, Good Health and Well-Being, and allows sharing and transferring of knowledge to fight this devastating disease around the world.

Three-dimensional magnetic resonance cholangiopancreatography for the diagnosis of biliary atresia in infants and neonates.

BACKGROUND AND OBJECTIVE: Magnetic resonance cholangiopancreatography (MRCP) is widely accepted for visualization of the biliary system. However, the sensitivity and specificity of MRCP for the diagnosis of biliary atresia (BA) are still not fully elucidated. This study aimed to investigate the diagnostic value of three-dimensional MRCP (3D-MRCP) for BA in a large cohort of cholestatic infants and neonates. METHODS: One hundred ninety patients with infant jaundice underwent 3D-MRCP and one or more of the following: (1) intraoperative cholangiography, (2) laparoscopic exploration and pathological examination, or/and (3) clinical therapy. Statistical analyses were performed to determine the diagnostic accuracy of 3D-MRCP for BA. RESULTS: Our study demonstrated that 158 of 190 patients were interpreted as having BA by 3D-MRCP; of those, 103 patients were confirmed as having BA, whereas 55 patients did not have BA. Of the 32 patients interpreted as non-BA cases by 3D-MRCP, one patient was misdiagnosed. The diagnostic accuracy for 3D-MRCP was 70.53% (134 of 190), the sensitivity was 99.04% (103 of 104), the specificity was 36.05% (31 of 86), the negative predictive value was 96.88% (31 of 32), the positive predictive value was 65.19% (103 of 158), the positive likelihood ratio was 2.7473, the negative likelihood ratio was 0.0267, and the Youden index was 0.3509. CONCLUSIONS: The sensitivity of 3D-MRCP in diagnosing BA was excellent, but the specificity was not as high as described in previous reports. 3D-MRCP can be an effective screening method but should be combined with other modalities to identify BA and distinguish it from other causes of infant jaundice.