An Enhanced Framework for Postpartum Depression Diagnosis

Postpartum depression (PPD) is a serious mental illness that affects women after childbirth. Identifying the illness accurately depends on the method that is used in diagnosis. As such, a model that accurately predicts PPD is needed to overcome the negative impacts of this illness on nursing mothers. Results obtained from human experts revealed further complexity issues in handling manual tools like the EPDS questionnaire by patients or caregivers. The dataset comprising symptoms that serve as indicators of the illness and expected outcomes was broken into dependent (y) and independent (x) variables respectively while following the Quasi-experimental methodology and tools like Python and Anaconda navigator open source package/libraries. The neural network algorithm was trained with 75% of the datasets (n=1500) obtained and then tested with the remaining 25%, this successfully yielded a 0.98% accuracy level and 100% prediction score for any PPD type. Given the subtleness, associated risk and complexity of the subject, more research is necessary to further explore other effective tools in artificial intelligence to tackle this health scare. Keywords: Postpartum Depression (PPD), Diagnosis, Symptoms, Mood Disorder, Pathogenesis etc. Journal Reference Format: Ogunjimi, O.L.A. (2025): ): An Enhanced Framework for Postpartum Depression Diagnosis. Social Informatics, Business, Politics, Law, Environmental Sciences & Technology Journal. Vol. 11, No. 3. Pp 1-16. www.isteams/socialinformaticsjournal. dx.doi.org/10.22624/AIMS/SIJ/V11N3P1