Curating a knowledge base for patients with neurosyphilis: a study protocol of a DEep learning Framework for pErsonalized prediction of Adverse prognosTic events in NeuroSyphilis (DEFEAT-NS)

Adverse prognostic events (APE) of neurosyphilis include ongoing syphilitic meningitis, meningovascular syphilis, parenchymatous neurosyphilis and death. Its complexity and rarity have the potential to result in the underestimated true burden of neurosyphilis worldwide, due to lack of recognition and under-reporting. The unmet need for a modern method of refined and targeted treatment of neurosyphilis is strengthened by the currently various distinct diagnostic criteria. The DEep learning Framework for pErsonalized prediction of Adverse prognosTic events in NeuroSyphilis study will develop and validate prediction models for personalised prediction of APE after initial diagnosis in neurosyphilis to aid shared decision-making and stratify care of patients with neurosyphilis at high risk of severe prognostic course.

We conducted formative research to conceptualise and design a robust and clinically acceptable deep learning framework. We will conduct a deep learning framework development and validation study using a retrospective, multicentre, longitudinal cohort design and applying unsupervised, semi-supervised machine learning and deep learning. It will be conducted following expert guidance for model development and validation and our previous research experience. This study design consists of six parts: development, calibration, validation, subgroup bias evaluation, clinical utility evaluation and explanation.

This study will be conducted according to the Declaration of Helsinki and the Harmonised Tripartite Guideline for Good Clinical Practice of the International Conference on Harmonisation. No patient will be directly involved in developing the study's research question, design and implementation. This study will be a retrospective analysis of already anonymised data; therefore, ethical approval and informed consent were waived by the institutional review board of School of Public Health (Shenzhen), Sun Yat-sen University. The results will be disseminated through a peer-reviewed publication.