Trial Outcomes & Findings for AI-based Medical Device Validation for Early Melanoma Detection (NCT NCT06221397)

NCT ID: NCT06221397

Last Updated: 2026-03-16

Results Overview

Measures the device's ability to distinguish between melanoma and non-melanoma cases using predicted probabilities.

Recruitment status

COMPLETED

Target enrollment

105 participants

Primary outcome timeframe

At the time of the single clinical visit (Baseline).

Results posted on

2026-03-16

Participant Flow

Recruitment took place at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto. Starting September 17, 2020, researchers used consecutive sampling to find adult patients with skin lesions suspected of malignancy. The process ended on November 13, 2023, after reaching a target ratio of melanoma cases. Investigators assessed eligibility during routine clinical visits.

After enrollment, the most significant event was a pilot data analysis of the first 40 participants. Researchers found that the initial cohort, which only included high-suspicion cases requiring biopsy, was not representative of daily clinical practice. Consequently, the intended sample was extended to 200 to include common benign lesions like nevi. But finally, 105 patients were enrolled. Two participants were excluded from the final analysis due to non-conclusive clinical diagnoses.

Participant milestones

Participant milestones
Measure
Patients With Suspected Cutaneous Malignancy
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Overall Study
STARTED
105
Overall Study
COMPLETED
103
Overall Study
NOT COMPLETED
2

Reasons for withdrawal

Reasons for withdrawal
Measure
Patients With Suspected Cutaneous Malignancy
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Overall Study
Two participants were excluded from the final analysis due to non-conclusive clinical diagnoses.
2

Baseline Characteristics

AI-based Medical Device Validation for Early Melanoma Detection

Baseline characteristics by cohort

Baseline characteristics by cohort
Measure
Patients With Suspected Cutaneous Malignancy
n=105 Participants
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Age, Continuous
62.1 years
STANDARD_DEVIATION 17.6 • n=10 Participants
Sex: Female, Male
Female
53 Participants
n=10 Participants
Sex: Female, Male
Male
52 Participants
n=10 Participants
Race (NIH/OMB)
American Indian or Alaska Native
0 Participants
n=10 Participants
Race (NIH/OMB)
Asian
0 Participants
n=10 Participants
Race (NIH/OMB)
Native Hawaiian or Other Pacific Islander
0 Participants
n=10 Participants
Race (NIH/OMB)
Black or African American
0 Participants
n=10 Participants
Race (NIH/OMB)
White
102 Participants
n=10 Participants
Race (NIH/OMB)
More than one race
3 Participants
n=10 Participants
Race (NIH/OMB)
Unknown or Not Reported
0 Participants
n=10 Participants
Region of Enrollment
Spain
105 Participants
n=10 Participants
Baseline Image Quality (DIQA Score)
71.4 Units on the DIQA Scale.
n=10 Participants

PRIMARY outcome

Timeframe: At the time of the single clinical visit (Baseline).

Population: This difference is due to the fact that a single participant could have more than one lesion; as a result, there were 105 patients but 114 lesions registered.

Measures the device's ability to distinguish between melanoma and non-melanoma cases using predicted probabilities.

Outcome measures

Outcome measures
Measure
Patients With Suspected Cutaneous Malignancy
n=114 Lesions
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Area Under the ROC Curve (AUC) for Melanoma Detection
0.8482 Probability area (value from 0 to 1)
Interval 0.7629 to 0.9222

PRIMARY outcome

Timeframe: At the time of the single clinical visit (Baseline)

Population: This difference is due to the fact that a single participant could have more than one lesion; as a result, there were 105 patients but 114 lesions registered.

Accuracy represents the percentage of all cases where the AI software's primary (top-ranked) prediction correctly matched the confirmed medical diagnosis. The "confirmed diagnosis" was determined by either a laboratory biopsy (the gold standard) or a consensus of expert dermatologists. To calculate this, the AI analyzed high-resolution dermoscopic images of skin lesions. The software succeeded if its highest-probability diagnosis category matched the actual disease category of the lesion. Only images meeting a minimum visual quality score (DIQA ≥ 5) were included in this analysis to ensure the results reflect performance in a professional clinical setting.

Outcome measures

Outcome measures
Measure
Patients With Suspected Cutaneous Malignancy
n=114 Lesions
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Accuracy for Melanoma Detection
0.81 Proportion of correct cases
Interval 0.65 to 0.94

PRIMARY outcome

Timeframe: At the time of the single clinical visit (Baseline).

Population: This difference is due to the fact that a single participant could have more than one lesion; as a result, there were 105 patients but 114 lesions registered.

The percentage of true positive melanoma cases correctly identified by the device.

Outcome measures

Outcome measures
Measure
Patients With Suspected Cutaneous Malignancy
n=114 Lesions
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Sensitivity for Melanoma Detection
0.93 Proportion of true positives
Interval 0.88 to 0.98

PRIMARY outcome

Timeframe: At the time of the single clinical visit (Baseline).

Population: This difference is due to the fact that a single participant could have more than one lesion; as a result, there were 105 patients but 114 lesions registered.

The percentage of true negative (benign) cases correctly identified by the device.

Outcome measures

Outcome measures
Measure
Patients With Suspected Cutaneous Malignancy
n=114 Lesions
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Specificity for Melanoma Detection
0.80 Proportion of true negatives
Interval 0.69 to 0.82

SECONDARY outcome

Timeframe: At the time of the single clinical visit (Baseline).

Population: This difference is due to the fact that a single participant could have more than one lesion; as a result, there were 105 patients but 114 lesions registered.

Evaluates if the correct diagnosis is within the Top-1 predictions across various skin disease categories (International Classification of Diseases).

Outcome measures

Outcome measures
Measure
Patients With Suspected Cutaneous Malignancy
n=114 Lesions
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Top-1 Accuracy for Multiple ICD Categories
0.55 Proportion of correct matches
Interval 0.45 to 0.65

SECONDARY outcome

Timeframe: At the time of the single clinical visit (Baseline).

Population: This difference is due to the fact that a single participant could have more than one lesion; as a result, there were 105 patients but 114 lesions registered.

Evaluates if the correct diagnosis is within the Top-3 predictions across various skin disease categories (International Classification of Diseases).

Outcome measures

Outcome measures
Measure
Patients With Suspected Cutaneous Malignancy
n=114 Lesions
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Top-3 Accuracy for Multiple ICD Categories
0.7569 Proportion of correct matches
Interval 0.67 to 0.83

SECONDARY outcome

Timeframe: At the time of the single clinical visit (Baseline).

Population: This difference is due to the fact that a single participant could have more than one lesion; as a result, there were 105 patients but 114 lesions registered.

Evaluates if the correct diagnosis is within the Top-5 predictions across various skin disease categories (International Classification of Diseases).

Outcome measures

Outcome measures
Measure
Patients With Suspected Cutaneous Malignancy
n=114 Lesions
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Top-5 Accuracy for Multiple ICD Categories
0.8422 Proportion of correct matches
Interval 0.76 to 0.9

SECONDARY outcome

Timeframe: At the time of the single clinical visit (Baseline).

Population: This difference is due to the fact that a single participant could have more than one lesion; as a result, there were 105 patients but 114 lesions registered.

Includes AUC, Sensitivity, and Specificity for detecting any malignant lesion (not limited to melanoma).

Outcome measures

Outcome measures
Measure
Patients With Suspected Cutaneous Malignancy
n=114 Lesions
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Area Under the ROC Curve (AUC) for Malignancy Detection
0.90 Probability area (value from 0 to 1)
Interval 0.84 to 0.94

SECONDARY outcome

Timeframe: At the time of the single clinical visit (Baseline).

Population: This difference is due to the fact that a single participant could have more than one lesion; as a result, there were 105 patients but 114 lesions registered.

The percentage of true positive malignant cases correctly identified by the device.

Outcome measures

Outcome measures
Measure
Patients With Suspected Cutaneous Malignancy
n=114 Lesions
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Sensitivity for Multiple Malignant Conditions Detection
0.81 Proportion of true positives
Interval 0.72 to 0.88

SECONDARY outcome

Timeframe: At the time of the single clinical visit (Baseline).

Population: This difference is due to the fact that a single participant could have more than one lesion; as a result, there were 105 patients but 114 lesions registered.

The percentage of true negative (benign) cases correctly identified by the device.

Outcome measures

Outcome measures
Measure
Patients With Suspected Cutaneous Malignancy
n=114 Lesions
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Specificity for Multiple Malignant Conditions Detection
0.86 Proportion of true negatives
Interval 0.77 to 0.94

SECONDARY outcome

Timeframe: At the time of the single clinical visit (Baseline).

Population: This difference is due to the fact that a single participant could have more than one lesion; as a result, there were 105 patients but 114 lesions registered.

Measures the Positive Predictive Value (PPV) and Negative Predictive Value (NPV) to determine the probability that a "malignant" or "benign" result from the device is correct.

Outcome measures

Outcome measures
Measure
Patients With Suspected Cutaneous Malignancy
n=114 Lesions
Group/Cohort Description The study group consists of adult patients (over 18 years old) who presented at the Dermatology Departments of Hospital Universitario Cruces and Hospital Universitario Basurto with skin lesions suspected of being malignant. As this is an observational study, participants were not assigned to any new medical interventions, drugs, or treatments as part of the research protocol.
Predictive Values (PPV and NPV) for Malignancy
Positive Predictive Value (PPV)
0.9247 Proportion of correct predictions
Interval 0.85 to 0.97
Predictive Values (PPV and NPV) for Malignancy
Negative Predictive Value (NPV)
0.6789 Proportion of correct predictions
Interval 0.54 to 0.81

Adverse Events

Patients With Suspected Cutaneous Malignancy

Serious events: 0 serious events
Other events: 0 other events
Deaths: 0 deaths

Serious adverse events

Adverse event data not reported

Other adverse events

Adverse event data not reported

Additional Information

Dr. Jordi Barrachina - Clinical Affairs Manager

AI Labs Group S.L.

Phone: +34 653 08 83 37

Results disclosure agreements

  • Principal investigator is a sponsor employee
  • Publication restrictions are in place