Over 81,000 patients assessed across the UK

Since 2020, Skin Analytics has assessed over 81,000 NHS patients for suspicion of skin cancer, helping Secondary Care organisations remove the need for more than 64% of face-to-face NHS urgent suspected skin cancer appointments.

Clinical performance over time

We have pioneered a market leading approach to performance monitoring and continual improvement for AI. We’re committed to providing the best technology and being completely transparent about our results.

This table is a summary of our performance since April 2022.

How do we set our performance targets?
Our performance targets are based on clinical performance published in the literature and agreed by our Clinical Advisory Committee which is composed of leading UK and international dermatologists and health economics experts.

Skin Analytics data up to Q4 2023 Post Market Surveillance Reports, with analysis based on 29,453 lesion outcomes.

For more detail please reach out to us directly.

Target

Current Version

(Released April 2022)

Melanoma

95%

SCC

95%

98%

BCC

90%

97%

All skin cancer

97%

Benign (biopsy and clinically confirmed)

71%

Benign (biopsy only)

29%

Research publications

Patient perspectives of artificial intelligence as a medical device in a skin cancer pathway

Authors: Kawsar A, Hussain K, Kalsi D, et al.

Journal: Frontiers in Medicine
Year: 2023

This study evaluated patient perspectives on DERM and found a positive sentiment towards its potential use. Patients felt confident in computers assisting doctors in diagnosis and management, as well as supporting general practitioners in assessing skin lesions. Patient acceptability is crucial for effective integration of AIaMD into care pathways, and continued feedback will refine services to ensure patient satisfaction.

Accuracy of an Artificial Intelligence as a medical device as part of a UK-based skin cancer teledermatology service

Authors: Helen Marsden, Chronis Kemos, Marcello Venzi, Mariana Noy, Shameera Maheswaran, Nicholas Francis, Christopher Hyde, Daniel Mullarkey, Dilraj Kalsi and Lucy Thomas

Journal: Frontiers in Medicine
Year: 2024

The primary objective of this study was to demonstrate that the AIaMD had a higher rate of correctly classifying lesions that did not need to be referred for biopsy or urgent face-to-face dermatologist review, compared to teledermatology standard of care, while achieving the same sensitivity to detect malignancy.

The AIaMD identified significantly more lesions that did not need to be referred for biopsy or urgent face-to-face dermatologist review, compared to teledermatologists. This has the potential to reduce the burden of unnecessary referrals when used as part of a teledermatology service.

Effectiveness of an image analyzing AI-based Digital Health Technology to identify Non-Melanoma Skin Cancer and other skin lesions: results of the DERM-003 study

Authors: Marsden H, Morgan C, Austin S, Degiovanni C, Venzi M, Kemos C, Greenhalgh J, Mullarkey D, Palamaras I.

Journal: Frontiers in Medicine
Year: 2023

This study across Royal Free NHS Foundation Trust, Poole Hospital NHS Foundation Trust and Newcastle Hospitals NHS Foundation Trust showed that DERM achieved high accuracy in diagnosing non-melanoma skin cancer, with potential to support timely diagnosis of malignant and premalignant skin lesions.

Real-world post-deployment performance of a novel machine learning-based digital health technology for skin lesion assessment and suggestions for post-market surveillance

Authors: Dilraj Kalsi, Lucy Thomas, Chris Hyde, Dan Mullarkey, Jack Greenhalgh, Justin M Ko

Journal: Frontiers in Medicine, Dermatology Section
Year: 2023

This paper reports prospective real-world DERM performance from AI skin cancer pathways at two National Health Service hospitals (UK) over more than a year. We show DERM performance in-line with sensitivity targets and pre-marketing authorisation research, and it reduced the caseload for hospital specialists in two pathways. Based on our experience we offer suggestions on key elements of post-market surveillance for AIaMDs.

Assessment of Accuracy of an Artificial Intelligence Algorithm to Detect Melanoma in Images of Skin Lesions

Authors: Phillips, M. et al.
Journal: Jama Network Open
Year: 2019

Skin Analytics conducted the world’s first prospective study for AI melanoma detection. Designed by leading clinicians we compared DERM to the performance of clinicians in a realistic hospital setting. Conducted across 7 NHS hospitals, led by the Royal Free London NHS Foundation Trust, the study demonstrated DERM’s performance in identifying melanoma.

Detection of Malignant Melanoma Using Artificial Intelligence: An Observational Study of Diagnostic Accuracy

Authors: Phillips, M., Greenhalgh, J., Marsden, H., Palamaras, I.
Journal: Dermatology Practical & Conceptual
Year: 2019

This study aimed to evaluate the accuracy of an artificial intelligence neural network (Deep Ensemble for Recognition of Melanoma [DERM]) to identify malignant melanoma from dermoscopic images of pigmented skin lesions and to show how this compared to doctors’ performance assessed by meta-analysis.

Current research with NHS Partners

In addition to the data we publish, much of which has been externally audited, Skin Analytics is working with a number of external partners to evaluate and support the expansion of our services across the NHS, including a number of independent and health economics evaluations.

Skin Analytics has completed a study with Chelsea and Westminster NHS Foundation Trust to review the impact that DERM could have when assessing all two week wait patients, including consideration of the health economic impact. We plan to publish this very soon.
Unity Insights has been appointed by the NHS AI Lab to independently evaluate Skin Analytics. Unity Insights have completed an interim analysis and are due to publish a detailed report later this year.
The University of Exeter are working to independently evaluate our recently granted project through the SBRI Cancer Programme that will focus on the use of DERM in populations who have not been referred by their GP.
Researchers at The University of Birmingham involved with the creation of the Medical Algorithmic Audit framework published in the Lancet and the BMJ journals have been applying this framework to a Skin Analytics deployment. Results are expected later this year and will be made available to the NHS.
Edge Health are working on behalf of the East Midlands Academic Health Science Network to evaluate our pathway at University Hospitals Leicester with a report due end 2023. You can see a case study of the pathway previously conducted here.

Health Economics

Since 2018, we have worked with teams from Health Enterprise East, the York Health Economics Consortium, Imperial Trust, as well as the Exeter Test Group to ensure that our services are truly sustainable for the NHS.

Today, we are uniquely well-placed to answer on the cost effectiveness of using DERM in secondary care, with over three years of comprehensive real-world evidence, derived from the assessment of more than 70,000 patients across 14 NHS sites.

Results from our health economic analysis show that Skin Analytics delivers robust NHS dermatology cost savings. Learn more.