Background and Objectives: We aimed to prospectively obtain data on pregnancies complicated with intrauterine growth restriction (IUGR) in the Prenatal Diagnosis Unit of the Emergency County Hospital of Craiova. We collected the demographic data of mothers, the prenatal ultrasound (US) features, the intrapartum data, and the immediate postnatal data of newborns. We aimed to assess the detection rates of IUGR fetuses (the performance of the US in estimating the actual neonatal birth weight), to describe the prenatal care pattern in our unit, and to establish predictors for the number of total hospitalization days needed postnatally. Materials and Methods: Data were collected from cases diagnosed with IUGR undergoing prenatal care in our hospital. We compared the percentile of estimated fetal weight (EFW) using the Hadlock 4 technique with the percentile of weight at birth. We retrospectively performed a regression analysis to correlate the variables predicting the number of hospitalization days. Results: Data on 111 women were processed during the period of 1 September 2019–1 September 2022. We confirmed the significant differences in US features between early- (Eo) and late-onset (Lo) IUGR cases. The detection rates were higher if the EFW was lower, and Eo-IUGR was associated with a higher number of US scans. We obtained a mathematical formula for estimating the total number of hospitalization days needed postnatally. Conclusion: Early- and late-onset IUGR have different US features prenatally and different postnatal outcomes. If the US EFW percentile is lower, a prenatal diagnosis is more likely to be made, and a closer follow-up is offered in our hospital. The total number of hospitalization days may be predicted using intrapartum and immediate postnatal data in both groups, having the potential to optimize the final financial costs and to organize the neonatal department efficiently.

Background and Objectives: In this study, we aimed to describe the clinical and ultrasound (US) features and the outcome in a group of patients suspected of or diagnosed with early onset intrauterine growth restriction (IUGR) requiring iatrogenic delivery before 32 weeks, having no structural or genetic fetal anomalies, managed in our unit. A secondary aim was to report the incidence of the condition in the population cared for in our hospital, data on immediate postnatal follow-up in these cases and to highlight the differences required in prenatal and postnatal care. Materials and Methods: We used as single criteria for defining the suspicion of early IUGR the sonographic estimation of fetal weight < p10 using the Hadlock 4 technique at any scan performed before 32 weeks’ gestation (WG). We used a cohort of patients having a normal evolution in pregnancy and uneventful vaginal births as controls. Data on pregnancy ultrasound, characteristics and neonatal outcomes were collected and analyzed. We hypothesized that the gestational age (GA) at delivery is related to the severity of the condition. Therefore, we performed a subanalysis in two subgroups, which were divided based on the GA at iatrogenic delivery (between 27+0 WG and 29+6 WG and 30+0–32+0 WG, respectively). Results: The prospective cohort study included 36 pregnancies. We had three cases of intrauterine fetal death (8.3%). The incidence was 1.98% in our population. We confirmed that severe cases (very early diagnosed and delivered) were associated with a higher number of prenatal visits and higher uterine arteries (UtA) pulsatility index (PI) centile in the third trimester—TT (compared with the early diagnosed and delivered). In the very early suspected IUGR subgroup, the newborns required significantly more NICU days and total hospitalization days. Conclusions: Patients with isolated very early and early IUGR—defined as ultrasound (US) estimation of fetal weight < p10 using the Hadlock 4 technique requiring iatrogenic delivery before 32 weeks’ gestation—require closer care prenatally and postnatally. These patients represent an economical burden for the health system, needing significantly longer hospitalization intervals, GA at birth and UtA PI centiles being related to it.

Introduction: This study aims to construct a predictive model that classifies the severity of coronavirus disease 2019 (COVID-19) in cancer patients by analyzing clinical, radiological, and demographic data. The goal is to assist healthcare providers in identifying high-risk patients and ensuring optimal resource allocation. Materials and methods: Data from 237 cancer patients diagnosed with COVID-19 was utilized to forecast disease severity based on various predictors. Key factors included cancer type and stage, intensive care unit admission, radiological assessments, ventilation status, obesity, and systemic inflammatory response syndrome (SIRS). The performance of multiple machine learning (ML) models, including Error-Correcting Output Codes (ECOC) frameworks built on Support Vector Machines (SVM), Decision Trees, K-Nearest Neighbors (KNN), Naive Bayes, Discriminant Analysis and an ensemble bagging method, was evaluated using 10-fold cross-validation. Model accuracy and receiver operating characteristic (ROC) curve scores served as the primary evaluation metrics. Results: KNN and ensemble bagging emerged as the most effective models, achieving accuracy rates of 100% and 98.3%, respectively, along with high area under the curve (AUC) values. These models excelled at identifying severe cases associated with intensive care unit (ICU) admission, ventilation, and metastatic cancer. Decision Trees demonstrated satisfactory performance with an accuracy of 82.55%, while SVM and Discriminant Analysis yielded moderate accuracy (64.26% and 65.11%). Naive Bayes underperformed, achieving only 40% accuracy, largely due to its assumption of feature independence. Conclusions: KNN and ensemble bagging models successfully predicted severe COVID-19 outcomes in cancer patients by capturing intricate relationships between factors such as ICU stays and ventilator support. Decision Trees also showed promise, but Naive Bayes was less reliable due to its simplified approach. The findings underscore the importance of non-linear models for predicting complex clinical outcomes. Despite robust evaluation through 10-fold cross-validation, the possibility of overfitting—particularly with KNN’s flawless accuracy—remains a concern. Additionally, the absence of external validation constrains the broader applicability of these findings to different patient populations.