Building a short video and audio database of cancer pain to help AI recognize patient pain
SENSAI: Seeing, hEaring, seNsing: Smart, Effortless and Objective Pain Assessment With Mobile AI Technology - DataBase Development
This project collects short face and voice recordings from adults with cancer who do and don't have pain to help test AI that tries to recognize pain from expressions.
Quick facts
| Study type | Observational |
|---|---|
| Enrollment | 200 (estimated) |
| Ages | 18 Years and up |
| Sex | All |
| Sponsor | Erasmus Medical Center Academic / other |
| Drugs / interventions | chemotherapy |
| Locations | 1 site (Rotterdam, South Holland) |
| Trial ID | NCT07262632 on ClinicalTrials.gov |
What this trial studies
This prospective, single-centre project at the Erasmus MC Cancer Institute will collect short (up to 60 seconds) video and audio recordings of adults with active cancer in two groups: patients admitted for cancer-related pain and controls admitted for chemotherapy with no pain. Participants will read a neutral prompt and, when relevant, briefly describe their pain while also completing a short questionnaire and allowing researchers to extract clinical data such as pain scores, medications, and tumour type. Recordings and linked clinical data will be securely stored under GDPR-compliant procedures and used to extract facial and vocal features with open-source tools like OpenFace and OpenSmile. The resulting labeled multimodal database is intended to support future development and testing of AI-based automatic pain recognition tools.
Who should consider this trial
Good fit: Adults (≥18) with active cancer who can communicate in Dutch or English and are admitted to Erasmus MC either for cancer-related pain or for chemotherapy with no pain are the intended participants.
Not a fit: Patients with conditions that impair facial expression or speech (for example facial paralysis, tracheostomy, severe speech impairment), those in critical or end-of-life care, or those experiencing non-cancer pain are unlikely to benefit or be eligible.
Why it matters
Potential benefit: If successful, this work could enable AI tools that detect pain from facial and vocal cues, helping clinicians identify and treat pain more quickly, especially when patients cannot communicate well.
How similar studies have performed: Facial and vocal feature–based approaches have shown promise in other pain and clinical settings, but applying these methods specifically to cancer-related pain is still novel and not yet validated.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: * Adult patients (≥18 years) * Diagnosed with cancer, active * Able to communicate verbally in Dutch or English * Able to provide written informed consent. * Pain group specific: * Experiencing pain related to cancer * Admitted to the hospital due to pain * Control group specific: * Not experiencing pain (NRS = 0) * Admitted to the hospital for chemotherapy Exclusion Criteria: * Cognitive, physical, or medical limitations that prevent participation in the audiovisual recording sessions or affect facial expressions or voice (e.g. facial paralysis, tracheostomy, severe speech impairment). * Critical illness or end-of-life care where participation would impose an additional burden. * Experiencing pain not associated with cancer * Infectious isolation precautions that prevent safe data collection
Where this trial is running
Rotterdam, South Holland
- Erasmus University Medical Centre — Rotterdam, South Holland, Netherlands (Recruiting)
Study contacts
- Principal investigator: Mark Mulder, dr.ir. — Erasmus University Medical Centre
- Study coordinator: Marsha Kamsteeg, MSc
- Email: m.kamsteeg@erasmusmc.nl
- Phone: +31 107040704
How to participate
- Review the eligibility criteria above with your treating physician.
- Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
- Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.