Artificial Intelligence (AI)-Based Prediction of Treatment Response in Sciatica Patients Receiving Transforaminal Epidural Steroid Injection

Artificial Intelligence-Based Predictive Models for Estimating Treatment Response in Sciatica Patients Treated With Transforaminal Epidural Steroid Injection (TFESI)

This prospective observational study aims to develop artificial intelligence (AI)-based predictive models to estimate treatment response in patients with sciatica undergoing transforaminal epidural steroid injection (TFESI). Approximately 1500 adult patients diagnosed with sciatica and scheduled for TFESI at Marmara University Training and Research Hospital Algology Clinic will be enrolled. Clinical, radiological, laboratory, and psychological data will be systematically collected before the procedure and at multiple follow-up points (1 hour, 3 weeks, 3 months, and 6 months post-treatment). Pain relief, measured by a 50% or greater reduction in the Numeric Rating Scale (NRS), will be the primary outcome. The study will use these comprehensive data sets to train and validate AI algorithms, aiming to create a decision-support system that can predict individual patient responses to TFESI and improve personalized treatment planning.

Inclusion Criteria:

• Age 18 years or older
• Presence of clinical and radiological radiculopathy findings consistent with L4, L5, or S1 nerve root involvement
• Symptoms present for at least 3 months
• Resistant to conservative treatment
• Scheduled for transforaminal epidural steroid injection (TFESI) by a pain medicine specialist
• Being willing to participate in the study and having signed a consent form

Exclusion Criteria:

• Malignancy
• Coagulation disorders
• Pregnancy
• History of allergic reaction to the injected substances
• History of major psychiatric disorders
• Presence of known coagulopathy
• Receiving systemic steroid treatment for any reason
• Having received an epidural steroid injection within the last 3 months