Symptoms timeline and outcomes in amyotrophic lateral sclerosis using artificial intelligence

Area of Science:

• Neuroscience
• Clinical Neurology
• Health Informatics

Background:

• Amyotrophic Lateral Sclerosis (ALS) is a fatal neurodegenerative motor neuron disease.
• Early diagnosis is critical for patient care and survival, yet patients often experience significant diagnostic delays.
• Understanding the clinical profile and diagnostic timeline in ALS is essential for improving patient outcomes.

Purpose of the Study:

• To describe the clinical profile of ALS patients using real-world data.
• To analyze the timing between symptom onset, diagnosis, and key clinical outcomes in ALS.
• To identify differences in clinical characteristics and disease progression between ALS subtypes.

Main Methods:

• Retrospective, multicenter study conducted in 5 hospitals and Primary Care services.
• Utilized Natural Language Processing (NLP) to extract clinical information from electronic health records (EHRs) of 250 ALS patients (2014-2018).
• Analyzed demographic data, symptom onset, diagnostic delays, and post-diagnosis interventions.

Main Results:

• The overall diagnostic delay from symptom onset was 11 months for most defining symptoms.
• Only 38.8% of patients saw a neurologist before diagnosis.
• Bulbar ALS patients showed more frequent and probable occurrence of gastrostomy, non-invasive ventilation, tracheostomy, and riluzole treatment compared to spinal ALS.

Conclusions:

• Significant diagnostic delays persist in Amyotrophic Lateral Sclerosis (ALS) care.
• Clinical characteristics and major disease-specific events differ between bulbar and spinal ALS subtypes.
• Natural Language Processing (NLP) demonstrates significant potential for analyzing rare neurological diseases within EHRs.