scholarly journals Identifying Elevated Risk for Future Pain Crises in Sickle-Cell Disease Using Photoplethysmogram Patterns Measured During Sleep: A Machine Learning Approach

2021 ◽  
Vol 3 ◽  
Author(s):  
Yunhua Ji ◽  
Patjanaporn Chalacheva ◽  
Carol L. Rosen ◽  
Michael R. DeBaun ◽  
Thomas D. Coates ◽  
...  
Keyword(s):  
Machine Learning ◽  
Sickle Cell Disease ◽  
Predictive Model ◽  
Sickle Cell ◽  
Microvascular Blood Flow ◽  
Term Therapy ◽  
Cell Disease ◽  
Sleep Study ◽  
Meta Model ◽  
Obstructive Sleep

Transient increases in peripheral vasoconstriction frequently occur in obstructive sleep apnea and periodic leg movement disorder, both of which are common in sickle cell disease (SCD). These events reduce microvascular blood flow and increase the likelihood of triggering painful vaso-occlusive crises (VOC) that are the hallmark of SCD. We recently reported a significant association between the magnitude of vasoconstriction, inferred from the finger photoplethysmogram (PPG) during sleep, and the frequency of future VOC in 212 children with SCD. In this study, we present an improved predictive model of VOC frequency by employing a two-level stacking machine learning (ML) model that incorporates detailed features extracted from the PPG signals in the same database. The first level contains seven different base ML algorithms predicting each subject's pain category based on the input PPG characteristics and other clinical information, while the second level is a meta model which uses the inputs to the first-level model along with the outputs of the base models to produce the final prediction. Model performance in predicting future VOC was significantly higher than in predicting VOC prior to each sleep study (F1-score of 0.43 vs. 0.35, p-value <0.0001), consistent with our hypothesis of a causal relationship between vasoconstriction and future pain incidence, rather than past pain leading to greater propensity for vasoconstriction. The model also performed much better than our previous conventional statistical model (F1 = 0.33), as well as all other algorithms that used only the base-models for predicting VOC without the second tier meta model. The modest F1 score of the present predictive model was due in part to the relatively small database with substantial imbalance (176:36) between low-pain and high-pain subjects, as well as other factors not captured by the sleep data alone. This report represents the first attempt ever to use non-invasive finger PPG measurements during sleep and a ML-based approach to predict increased propensity for VOC crises in SCD. The promising results suggest the future possibility of embedding an improved version of this model in a low-cost wearable system to assist clinicians in managing long-term therapy for SCD patients.

Screening for Obstructive Sleep Apnea in Children With Sickle Cell Disease: A Pilot Study

2020 ◽  
Author(s):  
Andrew J. Maroda ◽  
Matthew N. Spence ◽  
Stephen R. Larson ◽  
Jeremie H. Estepp ◽  
M. Boyd Gillespie ◽  
...  
Keyword(s):  
Obstructive Sleep Apnea ◽  
Sleep Apnea ◽  
Pilot Study ◽  
Sickle Cell Disease ◽  
Sickle Cell ◽  
Cell Disease ◽  
Obstructive Sleep

scholarly journals Contrast-enhanced ultrasound detects changes in microvascular blood flow in adults with sickle cell disease

PLoS ONE ◽  
2019 ◽  
Vol 14 (7) ◽  
pp. e0218783 ◽  
Author(s):  
Jonathan R. Lindner ◽  
Todd Belcik ◽  
Michael Widlansky ◽  
Leanne M. Harmann ◽  
Matthew S. Karafin ◽  
...  
Keyword(s):  
Blood Flow ◽  
Sickle Cell Disease ◽  
Sickle Cell ◽  
Microvascular Blood Flow ◽  
Contrast Enhanced Ultrasound ◽  
Cell Disease ◽  
Contrast Enhanced

Sickle Cell Disease Is a Risk Factor for Obstructive Sleep Apnea

CHEST Journal ◽  
2010 ◽  
Vol 138 (4) ◽  
pp. 322A
Author(s):  
JoAnn Eng ◽  
George A. Apergis ◽  
Samir Fahmy
Keyword(s):  
Obstructive Sleep Apnea ◽  
Risk Factor ◽  
Sleep Apnea ◽  
Sickle Cell Disease ◽  
Sickle Cell ◽  
Cell Disease ◽  
Obstructive Sleep

Adverse Events in Sickle Cell Patients with Nocturnal Hypoxia.

Blood ◽  
2005 ◽  
Vol 106 (11) ◽  
pp. 3793-3793
Author(s):  
Andrew D. Campbell ◽  
Megumi Okumura ◽  
Ndidi Unaka ◽  
Sally Hutchinson ◽  
Onyinye Onyekwere
Keyword(s):  
Obstructive Sleep Apnea ◽  
Sleep Apnea ◽  
Sickle Cell Disease ◽  
Adverse Events ◽  
Sickle Cell ◽  
Case Series ◽  
Acute Chest Syndrome ◽  
Cell Disease ◽  
Obstructive Sleep ◽  
The Relationship

Abstract The relationship between hemoglobin dexoygenation and sickling is well known. However, the relationship between hypoxia and severity of disease in sickle cell patients has not been well established. Recently, nocturnal hypoxemia has been associated with higher incidence of CNS events including strokes, and elevated TCDs. We present our case series on 13 patients(12 SS, 1 SC) with sickle cell disease (SCD) who have nocturnal hypoxia. Approximately 75 patients were screened at the University of Michigan Sickle Cell clinic for nocturnal hypoxia either by symptoms of obstructive sleep apnea or by longitudinal baseline clinic 02 saturations (02 Sat <92%). Of the 13 hypoxic pts, median baseline O2 Sat 90%(n=13, mean 90) and the median nocturnal O2sat (Nctnl 02 sat) 84%(n=13, mean 80%) with 10/13 with moderate-severe nocturnal hypoxia (O2sats<85%)based on sleep studies. Multiple adverse events noted in the cohort were pulmonary hypertension (PHTN TRJV>2.5, n=9, median 2.74/mean 2.74,) frequent pain episodes(>3visits to ER or hospitalizations/year, n=7, with 5 pts >5/year ), recurrent acute chest syndrome( ≥ 3 episodes, n= 10), CNS events (n=3 silent infarcts, vascular stenosis), priapism( (n=4, among 6 males ). Also reported were possible causes of the underlying hypoxia including obstructive sleep apnea(OSA)(n=7 of 11 pts), asthma(n= 10 of 13 pts), and chronic lung disease( n=8). In conclusion, the persistence of nocturnal hypoxia in pediatric sickle cell disease could possibly contribute to the development of severe complications of sickle cell disease. Treatment of underlying hypoxia (ie nighttime oxygen, maximize asthma treatment, T&A for OSA)may help prevent complications and lead to the improvement clinical symptoms. Further, chronic nocturnal hypoxia may complicate pulmonary disease and accelerated the development of PHTN. More studies are needed to clarify the mechanism of hypoxia in SCD. Table I. Clinical &Demographic Data of 13 SCD Patients with Nocturnal Hypoxia. Age:(6–22y/o, mean 15) Sex: M=6 F=7 Clinical: Total Mild Mod Sev. Genotype: SS=12, SC=1 Mean Range Nctnl Hypoxia(<%) 13 3 5(<85) 5(<80) Baseline O2 Sat(%) 90 +3.0 86–97 Obstr Sleep Apnea 7 3 3 1 Nctnl 02 sat (%) 80+8.4 59–87 Pulm HTN 9 4 4 1 Total #Apneic Events(11) 65.6+80 6–256 Rest. Lung Ds. 8 2 5 1 Obstr. Apneic Events(7) 27+68.5 0–221 # of Episodes <3 3–4 >4 Hypopneic Events(9) 32.5+38 0–132 ACS 2 5 5 TRJet Velocity 2.74+.42 2–3.5 Severe Pain Crises/yr 1 2 5

scholarly journals Contrast-Enhanced Ultrasound Detects Differences in Microvascular Blood Flow in Adults with Sickle Cell Disease Administered Regadenoson

Blood ◽  
2014 ◽  
Vol 124 (21) ◽  
pp. 2705-2705
Author(s):  
Jonathan R. Lindner ◽  
Michael Widlansky ◽  
Melinda D. Wu ◽  
Jillian Dargatz ◽  
Leanne M. Harmann ◽  
...  
Keyword(s):  
Skeletal Muscle ◽  
Blood Flow ◽  
Steady State ◽  
Sickle Cell Disease ◽  
Sickle Cell ◽  
Microvascular Blood Flow ◽  
Study Cohort ◽  
Cell Disease ◽  
Non Invasive ◽  
Contrast Enhanced

Abstract Background: Outcome measures for therapeutic studies in patients with sickle cell disease (SCD) are poor.Abnormal microvascular blood flow (MBF), the basis for tissue ischemia and injury associated with vaso-occlusion, would be an optimal outcome measure for SCD studies. Ideally, a modality to measure blood flow in SCD would non-invasively quantify microvascular tissue perfusion rather than assess conduit arterial flow through large vessels. Limitations of existing techniques to measure blood flow prevent their widespread use in clinical trials of patients with SCD. Contrast-enhanced ultrasound (CEU) is a non-invasive and portable technique that uses standard ultrasound equipment to measure microvascular perfusion and functional capillary patency. The primary objective of this study was to determine whether CEU is able to detect differences in the MBF of skeletal muscle: 1) before and after infusion with the adenosine A2A receptor (A2AR) agonist regadenoson, and 2) between steady state and vaso-occlusive crisis (VOC). Methods: CEU measurements were obtained in forearm skeletal muscle in adult HbSS patients. Two measures are used to calculate MBF: 1) velocity of RBCs through capillaries and 2) volume of blood perfused in an area of tissue. MBF is the product of RBC velocity and volume of blood. In one study cohort, MBF was measured in steady-state patients during a 24-hour infusion of regadenoson (1.44 µg/kg/hour). CEU perfusion imaging was obtained at baseline, 6 and 24 hours after initiation of regadenoson. In the second study cohort, CEU measurements were obtained within the same patient during a hospital admission for VOC and at steady state. MBF was expressed in terms of a ratio to baseline flow (pre-regadenoson) in cohort 1 and as a ratio of VOC to steady-state flow for cohort 2. Results: CEU measurements were obtained in13 patients administered regadenoson, and 7 patients at steady state and during VOC. Median age (range) of all patients studied was 24 years (20-45) and 55% were female. During regadenoson infusion, there was a median increase in skeletal muscle MBF of 29% at 6 hours (ratio 1.29, IQR 0.81) and 9% at 24 hours (ratio 1.09, IQR 1.40). Increase in MBF during regadenoson administration was largely due to higher RBC velocity (6 hours ratio: 1.24, IQR 0.88; 24 hours: ratio 1.12 IQR 0.85). There was a median decrease of 40% in skeletal muscle blood flow during VOC compared to steady state (ratio 0.60, IQR 0.27). Similarly, a decrease in RBC velocity accounted for most of the reduction in MBF in VOC compared to steady state (ratio 0.63, IQR 0.35). Conclusion: CEU measures of skeletal muscle MBF increased during a 24-hour infusion of regadenoson and decreased in VOC compared to steady state. Changes in RBC velocity, as opposed to the volume of blood perfused, accounted for most of the differences in MBF seen during regadenoson infusion and VOC. Alterations in rheology or vascular tone could explain these changes. These data provide additional evidence for the A2AR agonist regadenoson as a therapeutic modality for patients with SCD and suggest that CEU is a valid measure of blood flow in VOC. Taken together, the findings of this preliminary study demonstrate that CEU, a non-invasive, portable technique to measure MBF, could be used as an objective outcome measure for therapeutic studies in SCD. Disclosures Field: NKTT: Consultancy, Research Funding. Off Label Use: IND for regadenoson for treatment of VOC in sickle cell disease.

scholarly journals Machine learning predicts early-onset acute organ failure in critically ill patients with sickle cell disease

2019 ◽  
Author(s):  
Akram Mohammed ◽  
Pradeep S. B. Podila ◽  
Robert L. Davis ◽  
Kenneth I. Ataga ◽  
Jane S. Hankins ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intensive Care ◽  
Sickle Cell Disease ◽  
Kidney Disease ◽  
Organ Failure ◽  
Organ Dysfunction ◽  
Sickle Cell ◽  
Machine Learning Techniques ◽  
Cell Disease ◽  
Learning Techniques

AbstractBackgroundSickle cell disease (SCD) is a genetic disorder of the red blood cells, resulting in multiple acute and chronic complications including pain episodes, stroke, and kidney disease. Patients with SCD develop chronic organ dysfunction, which may progress to organ failure during disease exacerbations. Early detection of acute physiological deterioration leading to organ failure is not always attainable. Machine learning techniques that allow for prediction of organ failure may enable earlier identification and treatment, and potentially reduce mortality. We tested the hypothesis that machine learning physiomarkers could predict the development of organ dysfunction in an adult sample of patients with SCD admitted to intensive care units.Methods and FindingsWe studied 63 sequential SCD patients with 163 patient encounters, mean age 33.0±11.0 years, admitted to intensive care units, some of whom (6.7%) had pre-existing cardiovascular or kidney disease. A subset of these patient encounters (37; 23%) met sequential organ failure assessment (SOFA) criteria. The site of organ failure included: central nervous system (32), cardiovascular (11), renal (10), liver (7), respiratory (5) and coagulation (2) systems. Most (81.5%) of the patient encounters who experienced organ failure had single organ failure. The other 126 SCD patient encounters served as controls. A set of signal processing features (such as fast fourier transform, energy, continuous wavelet transform, etc.) derived from heart rate, blood pressure, and respiratory rate were identified to distinguish patients with SCD who developed acute physiological deterioration leading to organ failure, from SCD patients who did not meet the criteria. A random forest model accurately predicted organ failure up to six hours prior to onset, with a five-fold cross-validation accuracy of 94.57% (average sensitivity and specificity of 90.24% and 98.9% respectively).ConclusionsThis study demonstrates the viability of using machine learning to predict acute physiological deterioration heralded organ failure among hospitalized adults with SCD. The discovery of salient physiomarkers through machine learning techniques has the potential to further accelerate the development and implementation of innovative care delivery protocols and strategies for medically vulnerable patients.

Sign in / Sign up

Export Citation Format

Share Document