Pilot study for the clinical validation of a medical device for the automatic triage in teledermatology (Legit.Health_clinical_VH_2025)

Study type

• Study topic:
  Medical device
• Study type:
  Non-interventional study
• Scope of the study:
  Effectiveness study (incl. comparative)

  Healthcare resource utilisation
• Data collection methods:
  Combined primary data collection and secondary use of data

Non-interventional study

• Non-interventional study design:
  Cross-sectional

Study drug and medical condition

• Additional medical condition(s):
  Skin conditions in primary care

Population studied

• Short description of the study population:
  In this study, images from patients with different skin conditions and referred from primary care to dermatology will be analysed. ; In this way, these patients were seen in primary care due to different skin conditions, and the primary care physician decided to refer them to dermatology. ; In this way, we will assess if the use of the algorithm considers that the referral was acceptable and correct or not.
• Age groups:
  • Paediatric Population (< 18 years)

    • Neonate

      • Preterm newborn infants (0 – 27 days)
      • Term newborn infants (0 – 27 days)
    • Infants and toddlers (28 days – 23 months)
    • Children (2 to < 12 years)
    • Adolescents (12 to < 18 years)
  • Adult and elderly population (≥18 years)

    • Adults (18 to < 65 years)

      • Adults (18 to < 46 years)
      • Adults (46 to < 65 years)
    • Elderly (≥ 65 years)

      • Adults (65 to < 75 years)
      • Adults (75 to < 85 years)
      • Adults (85 years and over)
• Estimated number of subjects:
  30000

Study design details

• Study design:
  Retrospective study to assess the utility of a dermatology teleconsultation prioritisation algorithm based on machine learning. ; Study sample: teleconsultations with images submitted from primary care (adult and paediatric) to the Dermatology Department between 13 July 2020 and December 2023.
• Main study objective:
  The impact of implementing Legit.Health on reducing the average waiting time for skin cancer patients will be measured. ; To do so, both the mean wait time and the number of cases experiencing delays will be calculated and compared with the projected wait times once the Legit.Health algorithm is in place. ; This will provide a clear assessment of the algorithm’s effectiveness in the early diagnosis and management of melanoma and other malignant conditions. The KPIs to be measured are:; ; Reduction in mean waiting time for skin cancer patients.; ; Sensitivity and specificity of Legit.Health in malignancy detection.; ; Number of false negatives by primary care physicians identified by Legit.Health. False negatives are those referred as non-urgent but later confirmed malignant, or reclassified as high priority upon subsequent dermatologist review.; ; Number of false positives by primary care physicians identified by Legit.Health. False positives are those referred as urgent but later deemed benign or low priority by the dermatologist.; ; Trends in the quality of submitted images over time, ensuring the average image quality remains consistently adequate.; ; Identification of non‑malignant conditions requiring priority attention (e.g., drug eruptions and extensive inflammatory dermatoses).;
• Setting:
  The study will be carried out in the Vall d'Hebron University Hospital. In order to assess the main objective and KPIs, images from patients referred from primary care to dermatology will be selected and analysed by the medical device. ; The study will be conducted for 6 months and will assess whether the derivations were necessary and acceptable. The comparator here will be the derivation criteria of the primary care physician, which will be compared with the derivation criteria of the algorithm. The images should be from patients diagnosed with dermatological diseases by primary care physicians and subsequently referred to a dermatologist.
• Comparators:
  In this study, the referral criteria and level of priority assigned by the primary care physician for referring the patient to dermatology will be compared with the priority assigned by the medical device, with the dermatologist assessing which of the two is correct or more accurate.
• Data analysis plan:
  To validate the device, a comparative analysis will be conducted between the observed wait times and the possible wait times after prioritisation/triage through Legit.Health. ; In addition, a concordance analysis will be performed between the artificial intelligence tool, dermatologists, and primary care physicians. ; To this end, appropriate statistical tests will be used depending on the nature of the variables, as well as comparisons or estimates as appropriate.