Deep Learning Model Using Continuous Skin Temperature Data Predicts Labor Onset.

Background: Changes in body temperature anticipate labor onset in numerous mammals, yet this concept has not been explored in humans. Methods: We evaluated patterns in continuous skin temperature data in 91 pregnant women using a wearable smart ring. Additionally, we collected daily steroid hormone samples leading up to labor in a subset of 28 pregnancies and analyzed relationships among hormones and body temperature trajectory. Finally, we developed a novel autoencoder long-short-term-memory (AE-LSTM) deep learning model to provide a daily estimation of days until labor onset. Results: Features of temperature change leading up to labor were associated with urinary hormones and labor type. Spontaneous labors exhibited greater estriol to α-pregnanediol ratio, as well as lower body temperature and more stable circadian rhythms compared to pregnancies that did not undergo spontaneous labor. Skin temperature data from 54 pregnancies that underwent spontaneous labor between 34 and 42 weeks of gestation were included in training the AE-LSTM model, and an additional 40 pregnancies that underwent artificial induction of labor or Cesarean without labor were used for further testing. The model was trained only on aggregate 5-minute skin temperature data starting at a gestational age of 240 until labor onset. During cross-validation AE-LSTM average error (true - predicted) dropped below 2 days at 8 days before labor, independent of gestational age. Labor onset windows were calculated from the AE-LSTM output using a probabilistic distribution of model error. For these windows AE-LSTM correctly predicted labor start for 79% of the spontaneous labors within a 4.6-day window at 7 days before true labor, and 7.4-day window at 10 days before true labor. Conclusion: Continuous skin temperature reflects progression toward labor and hormonal status during pregnancy. Deep learning using continuous temperature may provide clinically valuable tools for pregnancy care.

Epicardial placement of human placental membrane protects from heart injury in a swine model of myocardial infarction.

Cardiac ischemic reperfusion injury (IRI) is paradoxically instigated by reestablishing blood-flow to ischemic myocardium typically from a myocardial infarction (MI). Although revascularization following MI remains the standard of care, effective strategies remain limited to prevent or attenuate IRI. We hypothesized that epicardial placement of human placental amnion/chorion (HPAC) grafts will protect against IRI. Using a clinically relevant model of IRI, swine were subjected to 45 min percutaneous ischemia followed with (MI + HPAC, n = 3) or without (MI only, n = 3) HPAC. Cardiac function was assessed by echocardiography, and regional punch biopsies were collected 14 days post-operatively. A deep phenotyping approach was implemented by using histological interrogation and incorporating global proteomics and transcriptomics in nonischemic, ischemic, and border zone biopsies. Our results established HPAC limited the extent of cardiac injury by 50% (11.0 ± 2.0% vs. 22.0 ± 3.0%, p = 0.039) and preserved ejection fraction in HPAC-treated swine (46.8 ± 2.7% vs. 35.8 ± 4.5%, p = 0.014). We present comprehensive transcriptome and proteome profiles of infarct (IZ), border (BZ), and remote (RZ) zone punch biopsies from swine myocardium during the proliferative cardiac repair phase 14 days post-MI. Both HPAC-treated and untreated tissues showed regional dynamic responses, whereas only HPAC-treated IZ revealed active immune and extracellular matrix remodeling. Decreased endoplasmic reticulum (ER)-dependent protein secretion and increased antiapoptotic and anti-inflammatory responses were measured in HPAC-treated biopsies. We provide quantitative evidence HPAC reduced cardiac injury from MI in a preclinical swine model, establishing a potential new therapeutic strategy for IRI. Minimizing the impact of MI remains a central clinical challenge. We present a new strategy to attenuate post-MI cardiac injury using HPAC in a swine model of IRI. Placement of HPAC membrane on the heart following MI minimizes ischemic damage, preserves cardiac function, and promotes anti-inflammatory signaling pathways.

An adaptable and non-invasive method for tracking Bifidobacterium animalis subspecies lactis 420 in the mouse gut.

Probiotic strains from the Bifidobacterium or Lactobacillus genera improve health outcomes in models of metabolic and cardiovascular disease. Yet, underlying mechanisms governing these improved health outcomes are rooted in the interaction of gut microbiota, intestinal interface, and probiotic strain. Central to defining the underlying mechanisms governing these improved health outcomes is the development of adaptable and non-invasive tools to study probiotic localization and colonization within the host gut microbiome. The objective of this study was to test labeling and tracking efficacy of Bifidobacterium animalis subspecies lactis 420 (B420) using a common clinical imaging agent, indocyanine green (ICG). ICG was an effective in situ labeling agent visualized in either intact mouse or excised gastrointestinal (GI) tract at different time intervals. Quantitative PCR was used to validate ICG visualization of B420, which also demonstrated that B420 transit time matched normal murine GI motility (~8 hours). Contrary to previous thoughts, B420 did not colonize any region of the GI tract whether following a single bolus or daily administration for up to 10 days. We conclude that ICG may provide a useful tool to visualize and track probiotic species such as B420 without implementing complex molecular and genetic tools. Proof-of-concept studies indicate that B420 did not colonize and establish residency align the murine GI tract.