Prediction of Post-Spinal Hypotension During Cesarean Delivery Using Machine Learning Algorithms
Prediction of Hypotension Following Spinal Anesthesia During Cesarean Delivery Using Machine Learning Algorithms: A Retrospective Training and Prospective Validation Study
This study aims to develop and prospectively validate machine learning models for predicting hypotension following spinal anesthesia in pregnant women undergoing elective cesarean delivery. The study consists of two cohorts: a retrospective training cohort including at least 200 patients and an independent prospective validation cohort including 50 patients.
Three machine learning algorithms (Random Forest, XGBoost, and Logistic Regression) will be trained using preoperative demographic, anthropometric, and baseline hemodynamic variables, including age, body mass index, parity, gestational age, baseline systolic blood pressure, and baseline heart rate.
The primary outcome is post-spinal hypotension, defined as a decrease in systolic blood pressure of at least 20% from baseline and/or the requirement for vasopressor administration within the first 15 minutes after spinal anesthesia.
The best-performing model will be evaluated in the independent prospective cohort to assess its predictive performance in a real clinical setting.
Studieoversikt
Status
Status
Forhold
Forhold
Studietype
Studietype
Registrering (Antatt)
Registrering
Kontakter og plasseringer
Studiekontakt
Studiekontakt
- Navn: ilke dolğun
- Telefonnummer: +905555485632
- E-post: ilkeser2004@gmail.com
Studiesteder
-
-
Merkez Mahallesi
-
Istanbul, Merkez Mahallesi, Tyrkia (Türkiye), 34250
- Istinye Üniversity
-
-
Deltakelseskriterier
Kvalifikasjonskriterier
Kvalifikasjonskriterier
Alder som er kvalifisert for studier
Tar imot friske frivillige
Kjønn som er kvalifisert for studier
Prøvetakingsmetode
Studiepopulasjon
The study population will consist of pregnant women aged 18-45 years with ASA physical status I-II, singleton pregnancies, and scheduled for elective cesarean delivery under spinal anesthesia. The study will include two cohorts: a retrospective training cohort of at least 200 eligible patients identified from hospital records between January 1, 2025, and October 1, 2026, and an independent prospective validation cohort of 50 eligible pregnant women. Participants will be assessed according to predefined inclusion and exclusion criteria. The study population will include women with no history of chronic hypertension or preeclampsia/eclampsia and without contraindications to neuraxial anesthesia.
Beskrivelse
Inclusion Criteria:
- Age between 18 and 45 years.
- ASA physical status I-II.
- Singleton pregnancy.
- Scheduled for elective cesarean delivery under spinal anesthesia.
Exclusion Criteria:
- Multiple pregnancy.
- Emergency cesarean delivery, including cases of fetal distress.
- Chronic hypertension.
- History of preeclampsia or eclampsia.
- Contraindications to neuraxial anesthesia.
Studieplan
Hvordan er studiet utformet?
Designdetaljer
- Observasjonsmodeller: COHORT
- Tidsperspektiver: OTHER
Hva måler studien?
Primære resultatmål
Primære resultatmål
Resultatmål |
Tiltaksbeskrivelse |
Tidsramme |
|---|---|---|
|
Predictive Performance of Machine Learning Models for Post-Spinal Hypotension
Tidsramme: Within the first 15 minutes following spinal anesthesia.
|
The predictive performance of machine learning models (Random Forest, XGBoost, and Logistic Regression) for identifying post-spinal hypotension will be evaluated using the area under the receiver operating characteristic curve (AUC-ROC). Post-spinal hypotension is defined as a decrease in systolic blood pressure of at least 20% from baseline and/or the requirement for vasopressor administration within the first 15 minutes following spinal anesthesia. Models will be developed using the retrospective training cohort and evaluated in the independent prospective validation cohort. |
Within the first 15 minutes following spinal anesthesia.
|
Samarbeidspartnere og etterforskere
Sponsor
Sponsor
Publikasjoner og nyttige lenker
Generelle publikasjoner
- Yang X, Li YM, Wang Q, Li R, Zhang P. Machine learning model based on RCA-PDCA nursing methods and differentiating factors to predict hypotension during cesarean section surgery. Comput Biol Med. 2024 May;174:108395. doi: 10.1016/j.compbiomed.2024.108395. Epub 2024 Apr 2.
- Hung KC, Liu CC, Huang YT, Wu JY, Chen JY, Ko CC, Lin CM, Hsing CH, Yew M, Chen IW. Predictive Efficacy of the Perfusion Index for Hypotension following Spinal Anesthesia in Parturient Undergoing Elective Cesarean Section: A Systematic Review and Meta-Analysis. Diagnostics (Basel). 2023 Aug 3;13(15):2584. doi: 10.3390/diagnostics13152584.
Studierekorddatoer
Studer hoveddatoer
Studiestart (Antatt)
Studiestart
Primær fullføring (Antatt)
Primær fullføring
Studiet fullført (Antatt)
Studiet fullført
Datoer for studieregistrering
Først innsendt
Først innsendt
Først innsendt som oppfylte QC-kriteriene
Først innsendt som oppfylte QC-kriteriene
Først lagt ut (Faktiske)
Først lagt ut
Oppdateringer av studieposter
Sist oppdatering lagt ut (Faktiske)
Sist oppdatering lagt ut
Siste oppdatering sendt inn som oppfylte QC-kriteriene
Siste oppdatering sendt inn som oppfylte QC-kriteriene
Sist bekreftet
Sist bekreftet
Mer informasjon
Begreper knyttet til denne studien
Andre studie-ID-numre
Andre studie-ID-numre
- HASEKI-CS-ML-2026
Legemiddel- og utstyrsinformasjon, studiedokumenter
Studerer et amerikansk FDA-regulert medikamentprodukt
Studerer et amerikansk FDA-regulert enhetsprodukt
Denne informasjonen ble hentet direkte fra nettstedet clinicaltrials.gov uten noen endringer. Hvis du har noen forespørsler om å endre, fjerne eller oppdatere studiedetaljene dine, vennligst kontakt register@clinicaltrials.gov. Så snart en endring er implementert på clinicaltrials.gov, vil denne også bli oppdatert automatisk på nettstedet vårt. .