Search

Background: Pregnancy triggers longitudinal metabolic alterations in women to allow precisely-programmed fetal growth. Comprehensive characterization of such a “metabolic clock” of pregnancy may provide a molecular reference in relation to studies of adverse pregnancy outcomes. However, a high-resolution temporal profile of metabolites along a healthy pregnancy remains to be defined.Methods: Two independent, normal pregnancy cohorts with high-density weekly urine sampling (discovery: 478 samples from 19 subjects at California; validation: 171 samples from 10 subjects at Alabama) were studied. Urine samples were profiled by liquid chromatography-mass spectrometry (LC-MS) for untargeted metabolomics, which was applied for gestational age dating and prediction of time to delivery.Results: 5,473 urinary metabolic features were identified. Partial least-squares discriminant analysis on features with robust signals (n = 1,716) revealed that the samples were distributed on the basis of the first two principal components according to their gestational age. Pathways of bile secretion, steroid hormone biosynthesis, pantohenate, and CoA biosynthesis, benzoate degradation, and phenylpropanoid biosynthesis were significantly regulated, which was collectively applied to discover and validate a predictive model that accurately captures the chronology of pregnancy. With six urine metabolites (acetylcholine, estriol-3-glucuronide, dehydroepiandrosterone sulfate, α-lactose, hydroxyexanoy-carnitine, and l-carnitine), models were constructed based on gradient-boosting decision trees to date gestational age in high accordance with ultrasound results, and to accurately predict time to delivery.Conclusion: Our study characterizes the weekly baseline profile of the human pregnancy metabolome, which provides a high-resolution molecular reference for future studies of adverse pregnancy outcomes.

Multisystem inflammatory syndrome in children (MIS-C) is a novel disease identified during the COVID-19 pandemic characterized by systemic inflammation following SARS-CoV-2 infection. Delays in diagnosing MIS-C may lead to more severe disease with cardiac dysfunction or death. Most pediatric patients recover fully with anti-inflammatory treatments, but early detection of MIS-C remains a challenge given its clinical similarities to Kawasaki disease (KD) and other acute childhood illnesses.We developed KIDMATCH ( K awasak I D isease vs M ultisystem Infl A mma T ory syndrome in CH ildren), a deep learning algorithm for screening patients for MIS-C, KD, or other febrile illness, using age, the five classical clinical KD signs, and 17 laboratory measurements prospectively collected within 24 hours of admission to the emergency department from 1448 patients diagnosed with KD or other febrile illness between January 1, 2009 and December 31, 2019 at Rady Children's Hospital. For MIS-C patients, the same data was collected from 131 patients between May 14, 2020 to June 18, 2021 at Rady Children's Hospital, Connecticut Children's Hospital, and Children's Hospital Los Angeles. We trained a two-stage model consisting of feedforward neural networks to distinguish between MIS-C and non MIS-C patients and then KD and other febrile illness. After internally validating the algorithm using 10-fold cross validation, we incorporated a conformal prediction framework to tag patients with erroneous data or distribution shifts, enhancing the model generalizability and confidence by flagging unfamiliar cases as indeterminate instead of making spurious predictions. We externally validated KIDMATCH on 175 MIS-C patients from 16 hospitals across the United States.KIDMATCH achieved a high median area under the curve in the 10-fold cross validation of 0.988 [IQR: 0.98-0.993] in the first stage and 0.96 [IQR: 0.956-0.972] in the second stage using thresholds set at 95% sensitivity to detect positive MIS-C and KD cases respectively during training. External validation of KIDMATCH on MIS-C patients correctly classified 76/83 (2 rejected) patients from the CHARMS consortium, 47/50 (1 rejected) patients from Boston Children's Hospital, and 36/42 (2 rejected) patients from Children's National Hospital.KIDMATCH has the potential to aid frontline clinicians with distinguishing between MIS-C, KD, and similar febrile illnesses in a timely manner to allow prompt treatment and prevent severe complications.Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Heart, Lung, and Blood Institute, Patient-Centered Outcomes Research Institute, National Library of Medicine.

Background: Recently, a resurgence of pertussis has been observed worldwide despite broad vaccination coverage. The purpose of this study was to identify the clinical characteristics and the aetiol...

Multisystem inflammatory syndrome in children (MIS-C) is a novel disease that was identified during the COVID-19 pandemic and is characterised by systemic inflammation following SARS-CoV-2 infection. Early detection of MIS-C is a challenge given its clinical similarities to Kawasaki disease and other acute febrile childhood illnesses. We aimed to develop and validate an artificial intelligence algorithm that can distinguish among MIS-C, Kawasaki disease, and other similar febrile illnesses and aid in the diagnosis of patients in the emergency department and acute care setting.In this retrospective model development and validation study, we developed a deep-learning algorithm called KIDMATCH (Kawasaki Disease vs Multisystem Inflammatory Syndrome in Children) using patient age, the five classic clinical Kawasaki disease signs, and 17 laboratory measurements. All features were prospectively collected at the time of initial evaluation from patients diagnosed with Kawasaki disease or other febrile illness between Jan 1, 2009, and Dec 31, 2019, at Rady Children's Hospital in San Diego (CA, USA). For patients with MIS-C, the same data were collected from patients between May 7, 2020, and July 20, 2021, at Rady Children's Hospital, Connecticut Children's Medical Center in Hartford (CT, USA), and Children's Hospital Los Angeles (CA, USA). We trained a two-stage model consisting of feedforward neural networks to distinguish between patients with MIS-C and those without and then those with Kawasaki disease and other febrile illnesses. After internally validating the algorithm using stratified tenfold cross-validation, we incorporated a conformal prediction framework to tag patients with erroneous data or distribution shifts. We finally externally validated KIDMATCH on patients with MIS-C enrolled between April 22, 2020, and July 21, 2021, from Boston Children's Hospital (MA, USA), Children's National Hospital (Washington, DC, USA), and the CHARMS Study Group consortium of 14 US hospitals.1517 patients diagnosed at Rady Children's Hospital between Jan 1, 2009, and June 7, 2021, with MIS-C (n=69), Kawasaki disease (n=775), or other febrile illnesses (n=673) were identified for internal validation, with an additional 16 patients with MIS-C included from Connecticut Children's Medical Center and 50 from Children's Hospital Los Angeles between May 7, 2020, and July 20, 2021. KIDMATCH achieved a median area under the receiver operating characteristic curve during internal validation of 98·8% (IQR 98·0-99·3) in the first stage and 96·0% (95·6-97·2) in the second stage. We externally validated KIDMATCH on 175 patients with MIS-C from Boston Children's Hospital (n=50), Children's National Hospital (n=42), and the CHARMS Study Group consortium of 14 US hospitals (n=83). External validation of KIDMATCH on patients with MIS-C correctly classified 76 of 81 patients (94% accuracy, two rejected by conformal prediction) from 14 hospitals in the CHARMS Study Group consortium, 47 of 49 patients (96% accuracy, one rejected by conformal prediction) from Boston Children's Hospital, and 36 of 40 patients (90% accuracy, two rejected by conformal prediction) from Children's National Hospital.KIDMATCH has the potential to aid front-line clinicians to distinguish between MIS-C, Kawasaki disease, and other similar febrile illnesses to allow prompt treatment and prevent severe complications.US Eunice Kennedy Shriver National Institute of Child Health and Human Development, US National Heart, Lung, and Blood Institute, US Patient-Centered Outcomes Research Institute, US National Library of Medicine, the McCance Foundation, and the Gordon and Marilyn Macklin Foundation.

Kawasaki disease (KD) is the leading cause of acquired heart disease in children. A key objective of research in KD is to reduce the risk of long-term cardiovascular sequelae by expediting timely diagnosis. The major challenge in KD diagnosis is that it shares clinical signs with other childhood febrile illnesses. Our previously single-center developed computer-based two-step algorithm was further tested by a five-center validation in US. This first blinded multi-center trial validated our approach, with sufficient sensitivity and positive predictive value, to identify most patients with KD diagnosed at centers across the US.We sought to determine if our algorithmic approach applied to an Asian cohort. This study involved 418 KD and 259 febrile controls (FC) from the Chang Gung Memorial Hospital in Taiwan. Our diagnostic algorithm retained sensitivity (379 of 418; 90.7%), specificity (223 of 259; 86.1%), PPV (379 of 409; 92.7%), and NPV (223 of 247; 90.3%) comparable to previous US 2016 single center and US 2020 fiver center results. Only 4.7% (15 of 418) of KD and 2.3% (6 of 259) of FC patients were identified as indeterminate. The algorithm identified 18 of 50 (36%) KD patients who presented 2 or 3 principal criteria. Of 418 KD patients, 157 were infants younger than one year and 89.2% (140 of 157) were classified correctly. Of the 44 patients with KD who had coronary artery abnormalities, our diagnostic algorithm correctly identified 43 (97.7%) including all patients with dilated coronary artery but one who found to resolve in 8 weeks.We assessed the performance of our KD diagnostic algorithm with a single center Asian cohort. This work demonstrates the applicability of our algorithmic approach and diagnostic portability, providing evidence to support the launch of an adequately powered, multicenter study for future Asian application in the emergency department setting. If deployed in Asia, our tool promises a cost-effective diagnostic approach to allow the timely management of Asian KD patients even in the absence of KD experts, to potentially enhance the outcome for KD patients and reduce the risk of coronary artery aneurysms.

Echocardiography-based screening for valvular disease in at-risk asymptomatic children can result in early diagnosis. These screening programs, however, are resource intensive and may not be feasible in many resource-limited settings. Automated echocardiographic diagnosis may enable more widespread echocardiographic screening, early diagnosis, and improved outcomes. In this feasibility study, the authors sought to build a machine learning model capable of identifying mitral regurgitation (MR) on echocardiography.Echocardiograms were labeled by clip for view and by frame for the presence of MR. The labeled data were used to build two convolutional neural networks to perform the stepwise tasks of classifying the clips (1) by view and (2) by the presence of any MR, including physiologic, in parasternal long-axis color Doppler views. The view classification model was developed using 66,330 frames, and model performance was evaluated using a hold-out testing data set with 45 echocardiograms (11,730 frames). The MR detection model was developed using 938 frames, and model performance was evaluated using a hold-out testing data set with 42 echocardiograms (182 frames). Metrics to evaluate model performance included accuracy, precision, recall, F1 score (average of precision and recall, ranging from 0 to 1, with 1 suggesting perfect precision and recall), and receiver operating characteristic analysis.For the parasternal long-axis view with color Doppler, the view classification convolutional neural network achieved an F1 score of 0.97. The MR detection convolutional neural network achieved testing accuracy of 0.86 and an area under the receiver operating characteristic curve of 0.91.A machine learning model is capable of discerning MR on transthoracic echocardiography. This is an encouraging step toward machine learning-based diagnosis of valvular heart disease on pediatric echocardiography.

ObjectiveThis study aimed to develop a blood test for the prediction of pre-eclampsia (PE) early in gestation. We hypothesised that the longitudinal measurements of circulating adipokines and sphingolipids in maternal serum over the course of pregnancy could identify novel prognostic biomarkers that are predictive of impending event of PE early in gestation.Study designRetrospective discovery and longitudinal confirmation.SettingMaternity units from two US hospitals.ParticipantsSix previously published studies of placental tissue (78 PE and 95 non-PE) were compiled for genomic discovery, maternal sera from 15 women (7 non-PE and 8 PE) enrolled at ProMedDx were used for sphingolipidomic discovery, and maternal sera from 40 women (20 non-PE and 20 PE) enrolled at Stanford University were used for longitudinal observation.Outcome measuresBiomarker candidates from discovery were longitudinally confirmed and compared in parallel to the ratio of placental growth factor (PlGF) and soluble fms-like tyrosine kinase (sFlt-1) using the same cohort. The datasets were generated by enzyme-linked immunosorbent and liquid chromatography-tandem mass spectrometric assays.ResultsOur discovery integrating genomic and sphingolipidomic analysis identified leptin (Lep) and ceramide (Cer) (d18:1/25:0) as novel biomarkers for early gestational assessment of PE. Our longitudinal observation revealed a marked elevation of Lep/Cer (d18:1/25:0) ratio in maternal serum at a median of 23 weeks’ gestation among women with impending PE as compared with women with uncomplicated pregnancy. The Lep/Cer (d18:1/25:0) ratio significantly outperformed the established sFlt-1/PlGF ratio in predicting impending event of PE with superior sensitivity (85% vs 20%) and area under curve (0.92 vs 0.52) from 5 to 25 weeks of gestation.ConclusionsOur study demonstrated the longitudinal measurement of maternal Lep/Cer (d18:1/25:0) ratio allows the non-invasive assessment of PE to identify pregnancy at high risk in early gestation, outperforming the established sFlt-1/PlGF ratio test.

Objective To evaluate whether longitudinal measurements of serological adipokines and sphingolipids can predict preeclampsia early in gestation. Design Retrospective multi-omics discovery and longitudinal validation. Setting Maternity units in two US hospitals. Methods A multi-omics approach integrating genomic and lipidomic discoveries was employed to identify leptin (Lep) and ceramide (Cer) as novel PE early gestational biomarkers. The levels of placental growth factor (PlGF), soluble fms-like tyrosine kinase (sFlt-1), Lep, and Cer in maternal sera were then determined by enzyme-linked immunosorbent and liquid chromatography-tandem mass spectrometric assays. Main outcome measures Interval from positive prediction to confirmative diagnosis. Results Genomic meta-analysis compiled six PE placental cohorts with 78 PE and 95 non-PE control placentas. The Testing Cohort included sera from 7 non-PE and 8 PE women collected at confirmatory diagnosis. The Validation Cohort included sera from 20 non-PE and 20 PE women collected longitudinally through gestation. Our findings revealed a marked elevation of maternal serum Leptin/Ceramide (d18:1/25:0) ratio from early gestation (a median of 23 weeks) when comparing later PE-complicated with uncomplicated pregnancies. The maternal Lep/Cer (d18:1/25:0) ratio significantly outperformed the established sFlt-1/PlGF ratio in predicting PE for sensitivity (85% vs. 40%), positive predictive value (89% vs. 42%), and AUC (0.92 vs. 0.52) from 5 to 25 weeks of gestation. Conclusions Non-invasive longitudinal assessment by serological evaluation of Lep/Cer (d18:1/25:0) ratio can case find early pregnancies at risk of preeclampsia, outperforming sFlt-1/PlGF ratio test. Tweetable abstract Non-invasive longitudinal assessment by serological evaluation of Lep and Cer ratio can predict preeclampsia early in gestation.

ObjectivesThe aim of this study was to develop a single blood test that could determine gestational age and estimate the risk of preterm birth by measuring serum metabolites. We hypothesised that serial metabolic modelling of serum analytes throughout pregnancy could be used to describe fetal gestational age and project preterm birth with a high degree of precision.Study designA retrospective cohort study.SettingTwo medical centres from the USA.ParticipantsThirty-six patients (20 full-term, 16 preterm) enrolled at Stanford University were used to develop gestational age and preterm birth risk algorithms, 22 patients (9 full-term, 13 preterm) enrolled at the University of Alabama were used to validate the algorithms.Outcome measuresMaternal blood was collected serially throughout pregnancy. Metabolic datasets were generated using mass spectrometry.ResultsA model to determine gestational age was developed (R2=0.98) and validated (R2=0.81). 66.7% of the estimates fell within ±1 week of ultrasound results during model validation. Significant disruptions from full-term pregnancy metabolic patterns were observed in preterm pregnancies (R2=−0.68). A separate algorithm to predict preterm birth was developed using a set of 10 metabolic pathways that resulted in an area under the curve of 0.96 and 0.92, a sensitivity of 0.88 and 0.86, and a specificity of 0.96 and 0.92 during development and validation testing, respectively.ConclusionsIn this study, metabolic profiling was used to develop and test a model for determining gestational age during full-term pregnancy progression, and to determine risk of preterm birth. With additional patient validation studies, these algorithms may be used to identify at-risk pregnancies prompting alterations in clinical care, and to gain biological insights into the pathophysiology of preterm birth. Metabolic pathway-based pregnancy modelling is a novel modality for investigation and clinical application development.

206 Background: Human serum metabolome profiles have been analyzed to explore the molecular changes that occur with aging. We hypothesized that deep metabolic profiling of sera with different ages would allow the identification of distinct metabolic chronologic patterns as a normal biological baseline to study personal aging. We further hypothesized that metabolic assessment of this chronologic deviation, resulting from advanced precancerous lesion (APL) and stage I/II/III CRC, from the normal reference baseline, would be instrumental for prognosis of relapse revealing underlying pathophysiology. Methods: A cohort of normal (n=3,616, training; n=1,170, testing), 631 advanced adenoma, 1,019 stage I, 404 stage II and 417 stage III serum samples were assembled. Innovative global LCMS metabolomic production were applied to deep profile these subjects. Identification of the age-associated molecular patterns in normal subjects, modeled with an elastic net algorithm, established the reference baseline to mirror a metabolic clock. CRC associated deviation from the precise chronologically paced metabolic patterns was quantified to associate the clinical endpoints of relapse, OSF and PFS, and to identify the tightly associated metabolic pathways. Results: We observed that for those CRCs, the predicted metabolic age can differ from the chronological age with consistent variations, resulting “older” or “younger” metabolic age subgroup in reference to the chronological age. Significant disruptions from the normal baseline were observed in CRCs patients, and consistent stage specific patterns were observed. Outlier, “Older” or “younger” metabolic age subgroup, CRC patients were found with significant future relapse enrichment. Predictive models were derived to case find the patients at risk of future relapse at the CRC diagnosis timepoint. Conclusions: Deviations from the meticulously timed metabolic aging patterns may provide utility to allow prognosis of future clinical endpoints of relapse and overall survival. Close examination of the underlying metabolic pathways, associated with CRC stage specific metabolic patterns, disrupting the baseline ageotypes, not only may improve the sensitivity and specificity of prognostic tests of CRC relapse, but also shed new insights into CRC therapeutics.

Recurrence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) positive detection in infected but recovered individuals has been reported. Patients who have recovered from coronavirus disease 2019 (COVID-19) could profoundly impact the health care system. We sought to define the kinetics and relevance of PCR-positive recurrence during recovery from acute COVID-19 to better understand risks for prolonged infectivity and reinfection. A series of 414 patients with confirmed SARS-Cov-2 infection, at The Second Affiliated Hospital of Southern University of Science and Technology in Shenzhen, China from January 11 to April 23, 2020. Statistical analyses were performed of the clinical, laboratory, radiologic image, medical treatment, and clinical course of admission/quarantine/readmission data, and a recurrence predictive algorithm was developed. 16.7% recovered patients with PCR positive recurring one to three times, despite being in strict quarantine. Younger patients with mild pulmonary respiratory syndrome had higher risk of PCR positivity recurrence. The recurrence prediction model had an area under the ROC curve of 0.786. This case series provides characteristics of patients with recurrent SARS-CoV-2 positivity. Use of a prediction algorithm may identify patients at high risk of recurrent SARS-CoV-2 positivity and help to establish protocols for health policy.

BACKGROUND Cardiac dysrhythmia is currently an extremely common disease. Severe arrhythmias often cause a series of complications, including congestive heart failure, fainting or syncope, stroke, and sudden death. OBJECTIVE The aim of this study was to predict incident arrhythmia prospectively within a 1-year period to provide early warning of impending arrhythmia. METHODS Retrospective (1,033,856 individuals enrolled between October 1, 2016, and October 1, 2017) and prospective (1,040,767 individuals enrolled between October 1, 2017, and October 1, 2018) cohorts were constructed from integrated electronic health records in Maine, United States. An ensemble learning workflow was built through multiple machine learning algorithms. Differentiating features, including acute and chronic diseases, procedures, health status, laboratory tests, prescriptions, clinical utilization indicators, and socioeconomic determinants, were compiled for incident arrhythmia assessment. The predictive model was retrospectively trained and calibrated using an isotonic regression method and was prospectively validated. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC). RESULTS The cardiac dysrhythmia case-finding algorithm (retrospective: AUROC 0.854; prospective: AUROC 0.827) stratified the population into 5 risk groups: 53.35% (555,233/1,040,767), 44.83% (466,594/1,040,767), 1.76% (18,290/1,040,767), 0.06% (623/1,040,767), and 0.003% (27/1,040,767) were in the very low-risk, low-risk, medium-risk, high-risk, and very high-risk groups, respectively; 51.85% (14/27) patients in the very high-risk subgroup were confirmed to have incident cardiac dysrhythmia within the subsequent 1 year. CONCLUSIONS Our case-finding algorithm is promising for prospectively predicting 1-year incident cardiac dysrhythmias in a general population, and we believe that our case-finding algorithm can serve as an early warning system to allow statewide population-level screening and surveillance to improve cardiac dysrhythmia care.

IMPORTANCEHow to appropriately care for patients who become PCR-negative for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is still not known. Patients who have recovered from coronavirus disease 2019 (COVID-19) could profoundly impact the health care system if a subset were to be PCR-positive again with reactivated SARS-CoV-2.OBJECTIVETo characterize a single center COVID-19 cohort with and without recurrence of PCR positivity, and develop an algorithm to identify patients at high risk of retest positivity after discharge to inform health care policy and case management decision-making.DESIGN, SETTING, AND PARTICIPANTSA cohort of 414 patients with confirmed SARS-CoV-2 infection, at The Second Affiliated Hospital of Southern University of Science and Technology in Shenzhen, China from January 11 to April 23, 2020.EXPOSURESPolymerase chain reaction (PCR) and IgM-IgG antibody confirmed SARS-CoV-2 infection.MAIN OUTCOMES AND MEASURESUnivariable and multivariable statistical analysis of the clinical, laboratory, radiologic image, medical treatment, and clinical course of admission/quarantine/readmission data to develop an algorithm to predict patients at risk of recurrence of PCR positivity.RESULTS16.7% (95CI: 13.0%-20.3%) patients retest PCR positive 1 to 3 times after discharge, despite being in strict quarantine. The driving factors in the recurrence prediction model included: age, BMI; lowest levels of the blood laboratory tests during hospitalization for cholinesterase, fibrinogen, albumin, prealbumin, calcium, eGFR, creatinine; highest levels of the blood laboratory tests during hospitalization for total bilirubin, lactate dehydrogenase, alkaline phosphatase; the first test results during hospitalization for partial pressure of oxygen, white blood cell and lymphocyte counts, blood procalcitonin; and the first test episodic Ct value and the lowest Ct value of the nasopharyngeal swab RT PCR results. Area under the ROC curve is 0.786.CONCLUSIONS AND RELEVANCEThis case series provides clinical characteristics of COVID-19 patients with recurrent PCR positivity, despite strict quarantine, at a 16.7% rate. Use of a recurrence prediction algorithm may identify patients at high risk of PCR retest positivity of SARS-CoV-2 and help modify COVID-19 case management and health policy approaches.Key PointsQuestionWhat are the characteristics, clinical presentations, and outcomes of COVID-19 patients with PCR retest positivity after resolution of the initial infection and consecutive negative tests? Can we identify recovered patients, prior to discharge, at risk of the recurrence of SARS-CoV-2 PCR positivity?FindingsIn this series of 414 COVID-19 inpatients discharged to a designated quarantine center, 69 retest positive (13 with 2 readmissions, and 3 with 3 readmissions). A multivariable model was developed to predict the risk of the recurrence of SARS-CoV-2 PCR positivity.MeaningRate and timing of the recurrence of PCR positivity following strict quarantine were characterized. Our prediction algorithm may have implications for COVID-19 clinical treatment, patient management, and health policy.

Ceramides and dihydroceramides are sphingolipids that present in abundance at the cellular membrane of eukaryotes. Although their metabolic dysregulation has been implicated in many diseases, our knowledge about circulating ceramide changes during the pregnancy remains limited. In this study, we present the development and validation of a high-throughput liquid chromatography-tandem mass spectrometric (LC/MS/MS) method for simultaneous quantification of 16 ceramides and 10 dihydroceramides in human serum within 5 mins by using stable isotope-labeled ceramides as internal standards (ISs). This method employs a protein precipitation method for high throughput sample preparation, reverse phase isocratic elusion for chromatographic separation, and Multiple Reaction Monitoring (MRM) for mass spectrometric detection. To qualify for clinical applications, our assay was validated against the FDA guidelines: the Lower Limit of Quantitation (LLOQ as low as 1 nM), linearity (R2>0.99), precision (Coefficient of Variation90%), stability (>85%), and carryover (

BACKGROUND: Recurrence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) positive detection in infected but recovered individuals has been reported. Patients who have recovered from coronavirus disease 2019 (COVID-19) could profoundly impact the health care system if a subset were to be polymerase chain reaction (PCR)-positive again with reactivated SARS-CoV-2. We sought to define the kinetics and relevance of PCR-positive recurrence during recovery from acute COVID-19 to better understand risks for prolonged infectivity and reinfection. METHODS: A series of COVID-19 414 patients, at The Second Affiliated Hospital of Southern University of Science and Technology in Shenzhen, China from January 11 to April 23, 2020. Univariable and multivariable statistical analysis of inpatient data were performed to develop an algorithm to predict patients at risk of recurrence of PCR positivity. FINDINGS: 16·7% recovered patients with PCR positive recurring one to three times, despite being in strict quarantine. Younger patients with mild pulmonary respiratory syndrome had higher risk of PCR positivity recurrence. The recurrence prediction model had an area under the ROC curve of 0·786. INTERPRETATION: This case series provides clinical characteristics of recovered COVID-19 patients with recurrent SARS-CoV-2 positivity, despite strict quarantine, at a 16·7% rate. Use of a recurrence prediction algorithm may identify patients at high risk of recurrent SARS-CoV-2 positivity and help understand reactivation and reinfection possibilities to establish protocols for health policy. FUNDING STATEMENT: This work was supported by grants from Sanming Project of Medicine in Shenzhen (Jia Huang, No. SZSM201812065); Bill & Melinda Gates Foundations (Lei Liu); and from National Natural Science Foundation of China (Jia Huang, No. 81501651) DECLARATION OF INTERESTS: The authors declare no competing interests. ETHICS APPROVAL STATEMENT: This study was approved by the Ethics Committee of the Second Affiliated Hospital of Southern University of Science and Technology. Written informed consent was obtained from all patients.

3613 Background: One-third of colorectal cancer (CRC) recurs following curative surgery and chemotherapy. Accordingly, novel methods are needed to predict recurrence to enable clinical course mitigating strategies. Serial monitoring of plasma by mass spectrometry (MS) and multi-omics modeling (MMO) of CRC relapse chronology provide the framework for liquid biopsy test development to supersede existing imaging modalities such as CT scans according to relapse related pathologies. We hypothesized that plasma MS and MMO analysis of relapse related pathologies can deconvolute high risk stratification for CRC recurrence within the cancer continuum of care pre/post-surgery and/or pre/post adjuvant chemotherapy (ACT). Methods: 189 CRC patients (Stage I-III) underwent one of three treatment modalities: Modality 1 (Surgery followed by ACT), Modality 2 (Surgery only), Modality 3 (Neoadjuvant chemotherapy followed by surgery and ACT). Plasma samples (n = 441) were collected from patients before surgery, 30 days post-op, and every 3 months until death or month 24 whichever came first. The MMO approach was used to analyze biological features encompassing native peptides, proteins, metabolites, lipids, and ceramides. MMO panels were developed comprising the significantly perturbed features as per the treatment modalities. These panels were used to predict relapse from plasma collected pre-op, 30-day post-op or after adjuvant chemotherapy. CEA levels were monitored in parallel. Results: Follow-up data was available for 135 patients (Stage I-III) and 25/135 had evidence of radiological recurrence. Irrespective of the treatment modality, longitudinal follow-up using the MMO panel was able to predict disease recurrence greater than 7 months before clinical progression was confirmed by CT scan. There was no significant correlation between longitudinal CEA levels and recurrence status, hence CEA levels alone did not provide any lead time advantage over the MMO panel or radiological surveillance. Kaplan-Meier (KM) survival analysis revealed that patients that were MMO panel positive had a poor survival irrespective of treatment modalities used: Modality 1 (HR = 6.2, p value = 0.003, test immediately post-surgery and immediately before ACT; HR = 31.6, p value = 0.01, test immediately after ACT); Modality 2 (HR = 11.2; p value = 0.01, test immediately after-surgery); Modality 3 (HR > 40, p value = 0.08, test immediately after neo-ACT and before-surgery; HR > 40, p value = 0.004, test immediately after-surgery). Conclusions: The MMO panel predicts CRC recurrence several months prior to detection by conventional CT scans, thus providing opportunity for alternative therapeutic strategies much earlier in the disease course.

Malignant gliomas remain incurable with a poor prognosis despite of aggressive treatment. We have been studying the development of brain tumors in a glioma rat model, where rats develop brain tumors after prenatal exposure to ethylnitrosourea (ENU), and there is a sizable interval between when the first pathological changes are noted and tumors become detectable with MRI. Our aim to define a molecular timeline through proteomic profiling of the cerebrospinal fluid (CSF) such that brain tumor commitment can be revealed earlier than at the presymptomatic stage. A comparative proteomic approach was applied to profile CSF collected serially either before, at and after the time MRI becomes positive. Elastic net (EN) based models were developed to infer the timeline of normal or tumor development respectively, mirroring a chronology of precisely timed, "clocked", adaptations. These CSF changes were later quantified by longitudinal entropy analyses of the EN predictive metric. False discovery rates (FDR) were computed to control the expected proportion of the EN models that are due to multiple hypothesis testing. Our ENU rat brain tumor dating EN model indicated that protein content in CSF is programmed even before tumor MRI detection. The findings of the precisely timed CSF tumor microenvironment changes at presymptomatic stages, deviation from the normal development timeline, may provide the groundwork for the understanding of adaptation of the brain environment in tumorigenesis to devise effective brain tumor management strategies.

PurposeTo develop a magneto-nanosensor (MNS) based multiplex assay to measure protein and autoantibody biomarkers from human serum for prostate cancer (CaP) diagnosis.Materials and methodsA 4-panel MNS autoantibody assay and a MNS protein assay were developed and optimized in our labs. Using these assays, serum concentration of six biomarkers including prostate-specific antigen (PSA) protein, free/total PSA ratio, as well as four autoantibodies against Parkinson disease 7 (PARK7), TAR DNA-binding protein 43 (TARDBP), Talin 1 (TLN1), and Caldesmon 1 (CALD1) and were analyzed. Human serum samples from 99 patients (50 with non-cancer and 49 with clinically localized CaP) were evaluated.ResultsThe MNS assay showed excellent performance characteristics and no cross-reactivity. All autoantibody assays showed a statistically significant difference between CaP and non-cancer samples except for PARK7. The most significant difference was the combination of the four autoantibodies as a panel in addition to the free/total PSA ratio. This combination had the highest area under the curve (AUC)- 0.916 in ROC analysis.ConclusionsOur results suggest that this autoantibody panel along with PSA and free PSA have potential to segregate patients without cancer from those with prostate cancer with higher sensitivity and specificity than PSA alone.

BackgroundThe clinical features of Kawasaki disease (KD) overlap with those of other paediatric febrile illnesses. A missed or delayed diagnosis increases the risk of coronary artery damage. Our computer algorithm for KD and febrile illness differentiation had a sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of 94.8%, 70.8%, 93.7% and 98.3%, respectively, in a single-centre validation study. We sought to determine the performance of this algorithm with febrile children from multiple institutions across the USA.MethodsWe used our previously published 18-variable panel that includes illness day, the five KD clinical criteria and readily available laboratory values. We applied this two-step algorithm using a linear discriminant analysis-based clinical model followed by a random forest-based algorithm to a cohort of 1059 acute KD and 282 febrile control patients from five children’s hospitals across the USA.ResultsThe algorithm correctly classified 970 of 1059 patients with KD and 163 of 282 febrile controls resulting in a sensitivity of 91.6%, specificity of 57.8% and PPV and NPV of 95.4% and 93.1%, respectively. The algorithm also correctly identified 218 of the 232 KD patients (94.0%) with abnormal echocardiograms.InterpretationThe expectation is that the predictive accuracy of the algorithm will be reduced in a real-world setting in which patients with KD are rare and febrile controls are common. However, the results of the current analysis suggest that this algorithm warrants a prospective, multicentre study to evaluate its potential utility as a physician support tool.

Background: The rapid deterioration observed in the condition of some hospitalized patients can be attributed to either disease progression or imperfect triage and level of care assignment after their admission. An early warning system (EWS) to identify patients at high risk of subsequent intrahospital death can be an effective tool for ensuring patient safety and quality of care and reducing avoidable harm and costs. Objective: The aim of this study was to prospectively validate a real-time EWS designed to predict patients at high risk of inpatient mortality during their hospital episodes. Methods: Data were collected from the system-wide electronic medical record (EMR) of two acute Berkshire Health System hospitals, comprising 54,246 inpatient admissions from January 1, 2015, to September 30, 2017, of which 2.30% (1248/54,246) resulted in intrahospital deaths. Multiple machine learning methods (linear and nonlinear) were explored and compared. The tree-based random forest method was selected to develop the predictive application for the intrahospital mortality assessment. After constructing the model, we prospectively validated the algorithms as a real-time inpatient EWS for mortality. Results: The EWS algorithm scored patients’ daily and long-term risk of inpatient mortality probability after admission and stratified them into distinct risk groups. In the prospective validation, the EWS prospectively attained a c-statistic of 0.884, where 99 encounters were captured in the highest risk group, 69% (68/99) of whom died during the episodes. It accurately predicted the possibility of death for the top 13.3% (34/255) of the patients at least 40.8 hours before death. Important clinical utilization features, together with coded diagnoses, vital signs, and laboratory test results were recognized as impactful predictors in the final EWS. Conclusions: In this study, we prospectively demonstrated the capability of the newly-designed EWS to monitor and alert clinicians about patients at high risk of in-hospital death in real time, thereby providing opportunities for timely interventions. This real-time EWS is able to assist clinical decision making and enable more actionable and effective individualized care for patients’ better health outcomes in target medical facilities.

Suicide is the tenth leading cause of death in the United States (US). An early-warning system (EWS) for suicide attempt could prove valuable for identifying those at risk of suicide attempts, and analyzing the contribution of repeated attempts to the risk of eventual death by suicide. In this study we sought to develop an EWS for high-risk suicide attempt patients through the development of a population-based risk stratification surveillance system. Advanced machine-learning algorithms and deep neural networks were utilized to build models with the data from electronic health records (EHRs). A final risk score was calculated for each individual and calibrated to indicate the probability of a suicide attempt in the following 1-year time period. Risk scores were subjected to individual-level analysis in order to aid in the interpretation of the results for health-care providers managing the at-risk cohorts. The 1-year suicide attempt risk model attained an area under the curve (AUC ROC) of 0.792 and 0.769 in the retrospective and prospective cohorts, respectively. The suicide attempt rate in the “very high risk” category was 60 times greater than the population baseline when tested in the prospective cohorts. Mental health disorders including depression, bipolar disorders and anxiety, along with substance abuse, impulse control disorders, clinical utilization indicators, and socioeconomic determinants were recognized as significant features associated with incident suicide attempt.

BACKGROUND The rapid deterioration observed in the condition of some hospitalized patients can be attributed to either disease progression or imperfect triage and level of care assignment after their admission. An early warning system (EWS) to identify patients at high risk of subsequent intrahospital death can be an effective tool for ensuring patient safety and quality of care and reducing avoidable harm and costs. OBJECTIVE The aim of this study was to prospectively validate a real-time EWS designed to predict patients at high risk of inpatient mortality during their hospital episodes. METHODS Data were collected from the system-wide electronic medical record (EMR) of two acute Berkshire Health System hospitals, comprising 54,246 inpatient admissions from January 1, 2015, to September 30, 2017, of which 2.30% (1248/54,246) resulted in intrahospital deaths. Multiple machine learning methods (linear and nonlinear) were explored and compared. The tree-based random forest method was selected to develop the predictive application for the intrahospital mortality assessment. After constructing the model, we prospectively validated the algorithms as a real-time inpatient EWS for mortality. RESULTS The EWS algorithm scored patients’ daily and long-term risk of inpatient mortality probability after admission and stratified them into distinct risk groups. In the prospective validation, the EWS prospectively attained a c-statistic of 0.884, where 99 encounters were captured in the highest risk group, 69% (68/99) of whom died during the episodes. It accurately predicted the possibility of death for the top 13.3% (34/255) of the patients at least 40.8 hours before death. Important clinical utilization features, together with coded diagnoses, vital signs, and laboratory test results were recognized as impactful predictors in the final EWS. CONCLUSIONS In this study, we prospectively demonstrated the capability of the newly-designed EWS to monitor and alert clinicians about patients at high risk of in-hospital death in real time, thereby providing opportunities for timely interventions. This real-time EWS is able to assist clinical decision making and enable more actionable and effective individualized care for patients’ better health outcomes in target medical facilities.

BACKGROUND Lung cancer is the leading cause of cancer death worldwide. Early detection of individuals at risk of lung cancer is critical to reduce the mortality rate. OBJECTIVE The aim of this study was to develop and validate a prospective risk prediction model to identify patients at risk of new incident lung cancer within the next 1 year in the general population. METHODS Data from individual patient electronic health records (EHRs) were extracted from the Maine Health Information Exchange network. The study population consisted of patients with at least one EHR between April 1, 2016, and March 31, 2018, who had no history of lung cancer. A retrospective cohort (N=873,598) and a prospective cohort (N=836,659) were formed for model construction and validation. An Extreme Gradient Boosting (XGBoost) algorithm was adopted to build the model. It assigned a score to each individual to quantify the probability of a new incident lung cancer diagnosis from October 1, 2016, to September 31, 2017. The model was trained with the clinical profile in the retrospective cohort from the preceding 6 months and validated with the prospective cohort to predict the risk of incident lung cancer from April 1, 2017, to March 31, 2018. RESULTS The model had an area under the curve (AUC) of 0.881 (95% CI 0.873-0.889) in the prospective cohort. Two thresholds of 0.0045 and 0.01 were applied to the predictive scores to stratify the population into low-, medium-, and high-risk categories. The incidence of lung cancer in the high-risk category (579/53,922, 1.07%) was 7.7 times higher than that in the overall cohort (1167/836,659, 0.14%). Age, a history of pulmonary diseases and other chronic diseases, medications for mental disorders, and social disparities were found to be associated with new incident lung cancer. CONCLUSIONS We retrospectively developed and prospectively validated an accurate risk prediction model of new incident lung cancer occurring in the next 1 year. Through statistical learning from the statewide EHR data in the preceding 6 months, our model was able to identify statewide high-risk patients, which will benefit the population health through establishment of preventive interventions or more intensive surveillance.

BACKGROUND Suicide is the 10th leading cause of death in the US. Over the past 15 years, the total suicide rate has increased 24% from 10.5 to 13.0 per 100,000 people. In Massachusetts the rate of death by suicide is three times the rate of homicide deaths. Approximately 60% of suicides die on the first attempt. Of the remaining 40% who survive the index attempt and receive emergency or hospital level of care, rates of subsequent completed suicide are exceptionally high, ranging from 2.3% to 4%. A recent study determined that risk factors for repeat suicide attempt and suicide differed, with alcohol use, younger age and cluster B personality disorders among the attempters and older age and alcohol use among the suicide completers. This data is from a small sample in Catalonia, Spain and whether it is generalizable to populations in the USA is yet to be determined. An early warning system (EWS) for suicide attempt could prove valuable for identifying those at risk of suicide attempts, and the contribution of repeated attempts to the risk of eventual death by suicide. OBJECTIVE In this study we sought to develop an early warning system (EWS) for high risk suicide attempt patients through the development of a population-based risk stratification surveillance system. Importantly, this EWS was designed to support case managers, primary care and mental health care practitioners participating in accountable care programs. The continuous use of the system in this program will help assess the ongoing EWS effectiveness. METHODS Data from individual patient electronic health records (EHRs) from the Berkshire Health System located in Pittsfield, MA. Advanced machine-learning algorithms and Deep Neural Networks were utilized in the process of feature selection and model building. A final risk score was calculated for each individual and calibrated to indicate the probability of a suicide attempt in the following one-year time period. Risk scores were subjected to individual level analysis in order to aid in the interpretation of the results for health care providers managing the at-risk cohorts. RESULTS The one-year suicide attempt risk model attained an area under the curve (AUC) of 0.792 and 0.769 in the retrospective and prospective cohorts, respectively. The suicide attempt rate in the “very high risk” category was 60 times greater than the population baseline when tested in the prospective cohorts, 10 times greater in the “high risk” group, and 5 times greater in the “medium risk” bin. Mental health disorders including depression, bipolar disorders and anxiety, along with substance abuse, impulse control disorders, clinical utilization indicators, and socio-economic determinants were recognized as significant features associated with incident suicide attempt. CONCLUSIONS Utilizing a single EHR dataset, an EWS for the determination of risk for suicide attempt was successfully developed and prospectively validated using deep learning modeling techniques.

Background Cardiac dysrhythmia is currently an extremely common disease. Severe arrhythmias often cause a series of complications, including congestive heart failure, fainting or syncope, stroke, and sudden death. Objective The aim of this study was to predict incident arrhythmia prospectively within a 1-year period to provide early warning of impending arrhythmia. Methods Retrospective (1,033,856 individuals enrolled between October 1, 2016, and October 1, 2017) and prospective (1,040,767 individuals enrolled between October 1, 2017, and October 1, 2018) cohorts were constructed from integrated electronic health records in Maine, United States. An ensemble learning workflow was built through multiple machine learning algorithms. Differentiating features, including acute and chronic diseases, procedures, health status, laboratory tests, prescriptions, clinical utilization indicators, and socioeconomic determinants, were compiled for incident arrhythmia assessment. The predictive model was retrospectively trained and calibrated using an isotonic regression method and was prospectively validated. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC). Results The cardiac dysrhythmia case-finding algorithm (retrospective: AUROC 0.854; prospective: AUROC 0.827) stratified the population into 5 risk groups: 53.35% (555,233/1,040,767), 44.83% (466,594/1,040,767), 1.76% (18,290/1,040,767), 0.06% (623/1,040,767), and 0.003% (27/1,040,767) were in the very low-risk, low-risk, medium-risk, high-risk, and very high-risk groups, respectively; 51.85% (14/27) patients in the very high-risk subgroup were confirmed to have incident cardiac dysrhythmia within the subsequent 1 year. Conclusions Our case-finding algorithm is promising for prospectively predicting 1-year incident cardiac dysrhythmias in a general population, and we believe that our case-finding algorithm can serve as an early warning system to allow statewide population-level screening and surveillance to improve cardiac dysrhythmia care.