Machine Learning Model for Classifying the Results of Fetal Cardiotocography Conducted in High-Risk Pregnancies.

PURPOSE: Fetal well-being is usually assessed via fetal heart rate (FHR) monitoring during the antepartum period. However, the interpretation of FHR is a complex and subjective process with low reliability. This study developed a machine learning model that can classify fetal cardiotocography results as normal or abnormal. MATERIALS AND METHODS: In total, 17492 fetal cardiotocography results were obtained from Ajou University Hospital and 100 fetal cardiotocography results from Czech Technical University and University Hospital in Brno. Board-certified physicians then reviewed the fetal cardiotocography results and labeled 1456 of them as gold-standard; these results were used to train and validate the model. The remaining results were used to validate the clinical effectiveness of the model with the actual outcome. RESULTS: In a test dataset, our model achieved an area under the receiver operating characteristic curve (AUROC) of 0.89 and area under the precision-recall curve (AUPRC) of 0.73 in an internal validation dataset. An average AUROC of 0.73 and average AUPRC of 0.40 were achieved in the external validation dataset. Fetus abnormality score, as calculated from the continuous fetal cardiotocography results, was significantly associated with actual clinical outcomes [intrauterine growth restriction: odds ratio, 3.626 (p=0.031); Apgar score 1 min: odds ratio, 9.523 (p<0.001), Apgar score 5 min: odds ratio, 11.49 (p=0.001), and fetal distress: odds ratio, 23.09 (p<0.001)]. CONCLUSION: The machine learning model developed in this study showed precision in classifying FHR signals. This suggests that the model can be applied to medical devices as a screening tool for monitoring fetal status.

The effect of chronic cerebral hypoperfusion on the pathology of Alzheimer's disease: A positron emission tomography study in rats.

Cerebrovascular disease is a potential risk factor for Alzheimer's disease (AD). Although acute cerebral hypoperfusion causes neuronal necrosis and infarction, chronic cerebral hypoperfusion induces apoptosis in neurons, but its effects on the cognitive impairment are not clear. The purpose of this study was to evaluate the effects of chronic cerebral hypoperfusion on AD pathology and cerebral glucose metabolism. A model of chronic cerebral hypoperfusion was established by ligating the common carotid arteries bilaterally in adult male rats (CAL group). Sham-operated rats underwent the same procedures without artery ligation (control group). At 12 weeks after ligation, expression levels of amyloid-ß (Aß) and hyperphosphorylated tau (p-tau), as well as the regional cerebral glucose metabolism, were evaluated using Western blots and positron emission tomography with fluorine-18 fluorodeoxyglucose, respectively. The expression levels of Aß in the frontal cortex and hippocampus and of p-tau in the temporal cortex were significantly higher in the CAL group than those in the control group. The cerebral glucose metabolism of the amygdala, entorhinal cortex, and hippocampus was significantly decreased in the CAL group compared to that in the control. These results suggest that chronic cerebral hypoperfusion can induce AD pathology and may play a significant role in AD development.

Risk factors for septic adverse events and their impact on survival in advanced ovarian cancer patients treated with neoadjuvant chemotherapy and interval debulking surgery.

OBJECTIVES: The aim of this study was to analyze risk factors for septic complications during adjuvant chemotherapy and their impact on survival in patients with advanced epithelial ovarian cancer treated with neoadjuvant chemotherapy (NACT) followed by interval debulking surgery (IDS). METHODS: We retrospectively reviewed the medical records of 69 patients with advanced epithelial ovarian cancer from 2004 to 2017. All patients underwent three cycles of NACT followed by IDS and adjuvant chemotherapy. We identified grade 3 or 4 hematologic complications and severe adverse events accompanied by neutropenia, including sepsis or septic shock, that occurred during treatment. Clinicopathologic data including demographic factors, preoperative medical conditions, surgical procedures, and survival times were evaluated. RESULTS: Of 69 patients, 27 (39.1%), 6 (8.8%), and 2 (2.9%) patients experienced grade 3 or 4 neutropenia, anemia, and thrombocytopenia, respectively, during NACT. Thirteen patients (18.8%) had a neutropenic fever with sepsis and 2 patients (2.9%) died of septic shock during adjuvant chemotherapy. Concurrent medical disease, splenectomy during IDS, and anemia or thrombocytopenia during NACT were significant risk factors for septic adverse events. In multivariate analysis, anemia (hemoglobin < 8 g/dL, p = 0.004) during NACT was the only significant factor associated with septic adverse events during adjuvant chemotherapy. Although there was no significant difference in progression-free survival, overall survival was significantly shorter in patients with septic adverse events (median, 82.3 vs. 17.3 months, p = 0.007). CONCLUSIONS: Grade 3 anemia during NACT may be an early indicator for septic adverse events during adjuvant chemotherapy. Considering the adverse impact on survival, scheme and dose of adjuvant chemotherapy should be tailored, and careful follow-up evaluation should be ensured in this patient group.

Relative bioavailability of a prototype oral solution of the Aurora A kinase inhibitor alisertib (MLN8237) in patients with advanced solid tumors.

OBJECTIVES: Alisertib (MLN8237) is an investigational, oral, small-molecule, selective inhibitor of Aurora A kinase. Phase I/II studies of powder-in-capsule (PIC) and enteric-coated tablet formulations of alisertib have determined the recommended phase II dose and have demonstrated anti-tumor activity. This phase I relative bioavailability study characterized the pharmacokinetics of a prototype oral solution (OS) of alisertib (developed for patients unable to swallow solid dosage forms) in reference to the PIC formulation in adult cancer patients. MATERIALS AND METHODS: A safety evaluation was undertaken first following a 3+3 design (OS starting dose, 15 mg). The relative bioavailability of alisertib OS vs. PIC was then evaluated following single dose administration of alisertib OS 25 mg and PIC 50 mg, using a 2-way crossover study design. RESULTS: The relative bioavailability (geometric mean dose-normalized AUCinf ratio) of alisertib OS vs. PIC formulation was 1.26 (90% confidence interval (CI): 1.09-1.47 (OS, n=17; PIC, n=18 evaluable patients)). These results support a distinguishable difference in bioavailability of alisertib between the two formulations (lower bound of 90% CI>1), with an estimated 26% higher total systemic exposure with alisertib OS vs. PIC. Alisertib absorption from OS was faster than from PIC, with a shorter median tmax (OS, 1 hour; PIC, 2 hours) and a geometric mean dose-normalized Cmax ratio (OS vs. PIC) of 1.90 (90% CI: 1.52 - 2.37). CONCLUSIONS: These findings inform the starting dose of alisertib OS to support further clinical evaluation of alisertib in patients unable to swallow solid dosage forms.