Ryu, Alexander J.

Background: Diagnosis of heart failure with preserved ejection fraction (HFpEF) is challenging due to non-diagnostic imaging outputs and discordant clinical features. Using only a single videoclip from a standard transthoracic echocardiogram (TTE), we developed an artificial intelligence (AI) model to differentiate patients with HFpEF from those without HFpEF. In a separate test population, we assessed the model’s ability to predict cardiac mortality. Methods: A 3-dimensional convolutional neural network was developed and trained on apical 4-chamber videoclips to classify patients with HFpEF (cases) versus without HFpEF (controls). Overall and cardiac mortality were obtained from the National Death Index. Results: The model was independently tested in 646 cases and 638 controls from 8 centers in 4 states. Sensitivity (87.8%; 84.5-90.9%) and specificity (81.9%; 78.2-85.6%) were high and sustained in various subgroups, including increased left atrial volume index, average E/e’ ≥15, and pulmonary artery systolic pressure >35 mm Hg. During follow-up (median [IQR] 3.4 [1.7-6.5] years), 540 patients died. There were 135 cardiac deaths attributable to heart failure (47%), coronary artery disease (41%), valve disease (4%), arrhythmia (4%), and others (5%). Compared to a negative output, all-cause mortality was higher in patients with a positive (HR 2.9 [2.4-3.5]) or uncertain output (HR 1.8 [1.2-2.5]), and cardiac mortality (Figure) was also higher in patients with a positive (HR 7.7 [4.5-13.3]) or uncertain output (HR 3.4 [1.4-8.1]). Cardiac mortality differences according to AI model prediction remained after age adjustment (positive: HR 5.9 [3.3-10.4], p<0.001; uncertain: HR 2.8 [1.2-6.8], p=0.02). Conclusion: A novel AI model to detect HFpEF based on a single routinely acquired TTE videoclip also identifies those at increased risk of cardiac mortality.

572 Background: The DEFeNDR (Decreasing the burden of Febrile Neutropenia through Dynamic RPM) program leverages in-home technology and centralized nurse monitoring to facilitate ambulatory management as an alternative to hospital-based care for cancer patients with neutropenic fever (NF). The objective of this study is to assess 30-day care utilization outcomes and patient satisfaction at 12 months after program implementation. Methods: Adult cancer patients hospitalized with NF were assessed daily for DEFeNDR program eligibility including MASCC score ≥21 (per ASCO guidelines) and no cellular therapy (CAR-T, BMT) in the prior 100 days. Patients agreeing to participate in the 30-day program received a cellular-enabled tablet with pre-connected devices for twice-daily vital signs and symptom assessments. Patient generated data were EHR-integrated in real-time. Embedded decision trees facilitated alerts for adverse trends. Virtual RPM nurses responded to alerts and utilized predefined care pathways. The primary endpoint was the mean proportion of evaluable days spent inpatient (mPEDI) within 30 days of eligibility comparing those who participated in the program (RPM) with those who declined (no RPM). Results: Of 85 eligible patients, 62 were offered DEFeNDR participation. Of those, 34 (54.8%) participated and 28 (45.2%) declined. Baseline characteristics and care utilization outcomes are summarized in Table 1. RPM patients experienced significantly lower mPEDI compared with unmonitored patients [5.1% (Standard Error [SE], 1.9%) vs 12.4% (SE, 2.1%); Wilcoxon Rank Sums p-value, 0.032]. RPM patients also experienced a trend toward fewer total hospital days and lower 30-day readmission rate, as well as shorter mean length of stay and lower rate of ICU utilization when readmitted. Non-visit care activities (EHR clinical communications, documentation, and patient messaging) were decreased from pre-hospital baseline for 58.8% of RPM patients versus 46.6% of those unmonitored. All RPM patients completing a patient satisfaction survey (n=9) indicated that they would recommend the program to patients with a similar condition. Conclusions: Cancer patients with NF participating in the DEFeNDR program experienced a significant reduction in mPEDI and care utilization compared with those who declined. Further studies are warranted to assess comparative effectiveness of this model vs usual care.[Table: see text]

Respiratory disorders, being one of the leading causes of disability worldwide, account for constant evolution in management technologies, resulting in the incorporation of artificial intelligence (AI) in the recording and analysis of lung sounds to aid diagnosis in clinical pulmonology practice. Although lung sound auscultation is a common clinical practice, its use in diagnosis is limited due to its high variability and subjectivity. We review the origin of lung sounds, various auscultation and processing methods over the years and their clinical applications to understand the potential for a lung sound auscultation and analysis device. Respiratory sounds result from the intra-pulmonary collision of molecules contained in the air, leading to turbulent flow and subsequent sound production. These sounds have been recorded via an electronic stethoscope and analyzed using back-propagation neural networks, wavelet transform models, Gaussian mixture models and recently with machine learning and deep learning models with possible use in asthma, COVID-19, asbestosis and interstitial lung disease. The purpose of this review was to summarize lung sound physiology, recording technologies and diagnostics methods using AI for digital pulmonology practice. Future research and development in recording and analyzing respiratory sounds in real time could revolutionize clinical practice for both the patients and the healthcare personnel.

Abstract Bowel sounds have been previously used to study intestinal motility and overall digestive health in various clinical settings. However, the blurred definition of bowel sounds and their subtypes, limited resources for interpretation, poor sensitivity, and low positive predictive value led to their restricted utility. Recent advances in artificial intelligence and machine learning have steered interest in developing unique tools using the phonoenterogram to analyze diverse bowel sounds. In our study, bowel sounds were recorded from eight healthy volunteers using the Eko Duo stethoscope. A novel deep-learning algorithm was designed to classify the recordings into baseline or prominent bowel sounds. A total of 11,210 data points (5,605 balanced sounds) were used to train and test the model to yield an accuracy of 0.895, a precision of 0.890, and a recall of 0.854 reflecting successful segregation of these sounds into respective groups. More extensive studies enrolling healthy and diseased subjects with a device specifically tailored to record bowel sounds are needed to generalize these results and determine their application in the patient population.

Abstract Blood pressure measurement in current medical practice relies on manual methods with the most widely used modality being sphygmomanometers. Utilizing the principle of Photoplethysmography, it is possible to provide an accurate reading of one’s blood pressure through light signals and photodetector devices. This research paper introduces a new Artificial Intelligence driven approach to predict Blood pressure levels and classify them according to the updated ACC (American College of Cardiology) criteria as Normal, Elevated, Stage I, and II Hypertension from the given PPG signal values using Machine Learning Models. This research paper aims to accurately read the Systolic and Diastolic Blood Pressure using Artificial Intelligence, place them into the correct value bins and further prove that the blood pressure values differ based on different skin tones in different light wavelengths such as red, infrared, and green. Machine Learning models such as the Support Vector Machine have shown an accuracy of 70.58% for Systolic Blood Pressure and Decision Tree with an accuracy of 74.4% for Diastolic Blood Pressure classification. The models used in this research are Support Vector Machine, Decision Tree and K-Nearest Neighbor. This research study has future applications and extensions to predict blood pressure levels for patients with different skin tones under different light radiations and PPG signal readings. Neural Network models will be developed to compare the blood predictions from this work.

Abstract Majority of hospitals still utilize manual methods for patient scheduling and predicting future appointments, resulting in longer wait times, hospital burnout and inadequate use of resources. A variety of avenues have been explored, including priority patient routing, tele-health, neural networks for improving ER efficiency, predicting no-shows, consultation duration variations, and optimizing operating room utilization. Addressing this issue, a study was conducted using 700 pre-visit notes of pancreatic patients to determine the requirement of endoscopic or biliary procedure. Through natural language processing and traditional or transfer learning algorithms, data could directly be sent to EPIC for nurses to assess in further decision making. Performance of the models was above average with the transfer learning method outperforming the traditional method. Although limited by less dataset and fewer circumstances to test the models on, the results exposed potential for future development with the possibility of patients reporting their chief concerns, in turn analyzed by algorithms, ultimately creating a smooth and effective patient itinerary.

1535 Background: Ambulatory management of select patients with febrile neutropenia (FN) may substantially decrease patients’ burden of illness and improve their quality of life, and is recommended by American Society of Clinical Oncology (ASCO) guidelines. However, in the United States, inpatient management remains the universal standard. We hereby report the health care utilization associated with the DEFeNDR remote patient monitoring (RPM) program, which leverages technology and virtual centralized nurse monitoring to facilitate ambulatory management of FN as an alternative to hospital-based care. Methods: Adult patients with cancer hospitalized with FN were assessed daily for eligibility to participate in the DEFeNDR program. Eligibility criteria included MASCC score >21 (per ASCO guidelines), and clinical judgment of the attending provider. Patients receiving cellular therapy (CAR-T, BMT) in the prior 100 days were excluded. Eligible patients agreeing to participate in the 30-day program were provided a kit containing cellular-enabled, pre-connected devices to facilitate monitoring, including twice-daily vital signs and electronic symptom questionnaires. Data were EHR-integrated in real-time and embedded decision trees facilitated alerts for adverse trends. RPM nurses responded to alerts and escalated care as indicated by pre-specified care pathways. Program assessment was performed at 20 weeks after full implementation. The primary outcome was the mean proportion of evaluable days spent inpatient (mPEDI) within 30 days of eligibility compared between those who participated in the program vs. those who declined. Results: 35 patients were offered participation in the DEFeNDR program. Of these, 17 participated and 18 declined. Baseline characteristics are provided. Participating patients experienced mPEDI of 6.9% (Standard Error [SE], 3.7%), compared with 17.7% (SE, 3.6%) for patients who declined participation. Due to small sample size, this difference (10.8%) did not reach statistical significance (Wilcoxon Rank Sums p-value, 0.15). 30-day mortality did not differ between groups. Mean inpatient days were also lower for patients participating in the program. Additionally, when readmitted, patients participating in RPM experienced a trend toward shorter length of stay and lower rates of intensive care unit utilization. Conclusions: A trend towards decreased burden of inpatient care is evident among patients participating in the DEFeNDR program. Further studies are warranted to assess comparative effectiveness of this model vs usual care.[Table: see text]

Production of bowel sounds, established in the 1900s, has limited application in existing patient-care regimes and diagnostic modalities. We review the physiology of bowel sound production, the developments in recording technologies and the clinical application in various scenarios, to understand the potential of a bowel sound recording and analysis device—the phonoenterogram in future gastroenterological practice. Bowel sound production depends on but is not entirely limited to the type of food consumed, amount of air ingested and the type of intestinal contractions. Recording technologies for extraction and analysis of these include the wavelet-based filtering, autoregressive moving average model, multivariate empirical mode decompression, radial basis function network, two-dimensional positional mapping, neural network model and acoustic biosensor technique. Prior studies evaluate the application of bowel sounds in conditions such as intestinal obstruction, acute appendicitis, large bowel disorders such as inflammatory bowel disease and bowel polyps, ascites, post-operative ileus, sepsis, irritable bowel syndrome, diabetes mellitus, neurodegenerative disorders such as Parkinson’s disease and neonatal conditions such as hypertrophic pyloric stenosis. Recording and analysis of bowel sounds using artificial intelligence is crucial for creating an accessible, inexpensive and safe device with a broad range of clinical applications. Microwave-based digital phonoenterography has huge potential for impacting GI practice and patient care.

TSLP is a newly‐identified cytokine thought to be involved in airway inflammation. TSLP levels have been shown to significantly increase in airway smooth muscle (ASM) with chronic obstructive pulmonary disease and other inflammatory diseases. This study evaluated the expression of TSLP in human ASM cells and tissues exposed to cigarette smoke extract (CSE) or to pro‐inflammatory cytokines TNF‐α and IL‐13. In parallel studies, TSLP expression was evaluated in house dust mite sensitized mice with or without cigarette smoke exposure. The presence of TSLP and TSLP‐receptors were verified using Western blot analysis. Exposure to TNF‐α and IL‐13 (50 ng/mL each) for 24 hours resulted in increased TSLP and TSLP‐R expression. Simultaneous exposure to functional TSLP antibody blunted cytokine induced TSLP‐R expression. Interestingly, acute CSE (30 min) down‐regulated TSLP expression, while chornic exposure increased expression. Similarly, [Ca 2+ ] i responses of human ASM to histamine were blunted with chronic CSE exposure, effects reversed by TSLP and TSLP‐R antibodies. Sensitized mice demonstrated an increase in TSLP expression in whole lung samples. This effect was even further enhanced in dust mite sensitized mice exposed to chronic cigarette smoke. These results indicate a complex role for TSLP in ASM inflammation with or without cigarette smoke exposure. Supported by NIH grants HL088029, HL090595, RR024150, and the Flight Attendants Medical Research Institute (FAMRI). Additional funding from the Department of Anesthesiology, Mayo Clinic.