Using machine learning for process improvement in sepsis management.

INTRODUCTION: In the U.S., sepsis afflicts 1.7 million adults, causing 270,000 deaths each year. Early detection of sepsis could decrease the number of deaths by 92,000 annually and decrease hospital expenditures by 1.5 billion USD. Few prior studies and reviews have presented a holistic understanding of the relationship between machine learning and existing process improvement measures. This study, in addition to discussing machine learning and existing process improvements measures, elaborates on the disadvantages and the barriers to integrating machine learning into the clinic. This article synthesizes previous studies to educate healthcare professionals on effectively managing sepsis by leveraging the benefits of machine learning. METHODS: This study used the PubMed database. Search terms include sepsis antibiotics, sepsis process improvement, sepsis machine learning. Our search criteria included previous studies published between January 1, 2017, and February 1, 2022. RESULTS/DISCUSSION: Although machine learning algorithms have better predictive capabilities, their effectiveness in the clinical setting is limited as studies show mixed results because the medical staff often fails to intervene. To overcome poor interventional response, clinicians need to work with the facility's IT department to ensure integration into clinical workflow and minimize alert-fatigue. Algorithms should enhance the productivity of clinical teams, not attempt to replace them entirely. CONCLUSION: Hospitals can employ process improvement measures that effectively utilize machine learning algorithms to ensure integration into clinical workflows. Healthcare professionals can utilize workflow tools in addition to the predictive capabilities of machine learning to enhance clinical decisions in sepsis.

Androgen regulation of signaling pathways in late fetal mouse lung development.

During lung development there is tension between positive and negative regulators of fibroblast-epithelial communication controlling type II cell differentiation. A clinical consequence of imbalance of this tension is the increased risk for respiratory distress syndrome in male infants. We hypothesized that chronic intrauterine androgen exposure alters fetal lung fibroblast maturation by down-regulating epidermal growth factor receptor (EGF-R) activity and by up-regulating transforming growth factor-beta receptor (TGFbeta-R) activity, leading to an inhibition of surfactant protein B (SP-B) and -C (SP-C) gene expression in type II cells. We treated pregnant mice with dihydrotestosterone (DHT; 2 mg/day) or vehicle for 7 days, starting on gestational day 11. On day 18, EGF binding, EGF-R phosphorylation, TGFbeta-R binding, and TGFbeta1-induced cell proliferation were studied in sex-specific fibroblast cultures. SP-B and -C messenger RNA levels were measured in whole lungs. Chronic DHT treatment reduced both EGF binding (females to 78+/-8% and males to 65+/-9% of controls) and EGF-induced EGF-R phosphorylation. TGFbeta-R binding was increased (females to 173+/-39% and males to 280+/-64% of controls), and TGFbeta-induced cell proliferation was increased in female cells (231+/-57% of controls). SP-B and -C messenger RNA expression was reduced to 55+/-10% and 75+/-4%, respectively. We conclude that chronic DHT exposure beginning early in lung development alters the balance of growth factor signaling that regulates lung maturation.

Effects of epidermal growth factor (EGF) on the development of EGF-receptor (EGF-R) binding in fetal rabbit lung organ culture.

Epidermal growth factor (EGF) causes gender- and development-specific changes in fetal lung surfactant synthesis. We hypothesized that the effects of EGF on development of surfactant synthesis are related to effects on EGF receptor (EGF-R) expression. We prepared sex-specific fetal rabbit lung organ cultures on gestational days 21 and 24 (term = 31 days) in Waymouth's medium + 10% charcoal-stripped fetal calf serum as control or with added EGF (10 ng/mL). After 3, 5, and 7 days of culture, we measured specific EGF-R binding in fetal lung plasma membrane preparations. Analysis of variance (ANOVA) revealed significant effects of fetal gender (P = 0.0003), time in culture (P = 0.01), and EGF treatment (P = 0. 0003) on EGF specific binding. In control cultures from days 21 and 24 (both male and female), EGF specific binding tended to decrease with time in culture. Specific binding in EGF-treated female 21-day cultures was significantly higher than in controls, both after 5 days (184% of control, P = 0.007) and after 7 days (151% of control, P = 0.01; Bonferroni multiple comparisons) of treatment, whereas males exhibited no response to EGF treatment. As opposed to these effects in 21-day cultures, EGF had little effect on 24-day cultures. We conclude that EGF affects the expression of the EGF-R on EGF specific binding in the fetal lung. The development of surfactant synthesis in the fetal lung may be controlled by upregulation of the EGF-R.

Transforming growth factor beta 1 binding and receptor kinetics in fetal mouse lung fibroblasts.

TGFbeta1 inhibits fetal lung maturation in vitro. As TGFbeta1 is present in fetal lung, mechanisms must exist to overcome this inhibition and allow late gestation maturation to progress. We studied the ontogeny of TGFbeta1 binding, and TGFbeta receptor kinetics and subtypes in primary cultures of fetal mouse lung fibroblasts from Day 16 to Day 18 of gestation. TGFbeta1 specific binding in fetal lung fibroblasts declined with advancing gestation. The decrease occurred earlier, and was more pronounced in female fibroblasts (50% decrease) than in the male fibroblasts (29% decrease). Dihydrotestosterone treatment of Day 18 female fibroblasts resulted in a dose-dependent increase in TGFbeta1 binding. Scatchard analysis revealed a decline in receptor number with advancing gestation (Day 16 female Bmax: 7.3 x 10(-16); Day 18 female Bmax: 5.5 x 10[-16]) whereas binding affinities remained constant. Affinity labeling followed by chemical cross-linking and autoradiography showed the three known TGFbeta receptor subtypes at both Days 16 and 18 of gestation. The relative abundance of nonsignaling Type III receptors in comparison to signaling Type II and Type I receptors was increased at Day 18 versus Day 16. Incorporation of [3H]thymidine into DNA after treatment with TGFbeta1 changed from Day 16 to Day 18, consistent with changes previously reported between fetal and adult lung fibroblasts. We conclude that as fetal mouse lung maturation progresses, TGFbeta receptor number decreases in fibroblasts, the relative proportion of nonsignaling versus signaling receptor types increases, and the response to TGFbeta1 stimulation changes. These changes may contribute to overcoming TGFbeta1 inhibition of lung maturation.