Skinive neural network uses a machine-learning algorithm to calculate the risk rating of skin pathologies. For this study, we used Skinive’s 2020 and 2021 versions trained on 64,000 and 115,000 images respectively. Three validation datasets were used to assess the sensitivity of the algorithm: precancer + cancer, HPV skin pathology, acne, containing 285 images in each set. The specificity has been calculated on a separate validation set containing 6,000 benign neoplasm cases.
The purpose of this study was to estimate the accuracy of Skinive’s algorithm.
We have improved the algorithm to show a statistically significant decrease in the number of neural network errors in determining the risks of skin pathologies.
The results of sensitivity and specificity of the Skinive neural network indicate that the algorithm is highly accurate in detecting various neoplasms and skin diseases. The sensitivity of the Skinive neural network in detecting malignant neoplasms was 89.1% and 95.4% in 2020 and 2021 respectively. The specificity of Skinive’s neural network in determining benign neoplasms was 95.3% in 2020 and 97.9% in 2021. For all skin neoplasms: in 2020, the sensitivity was 95.3%, for specificity 93.5%; in 2021, it was 97.9% and 97.1% respectively.
Summing up Skinive’s 2021 activities, we are excited to announce the list of events and conferences Skinive participated in during Autumn 2021:
1. The Next Web Conference (TNW) (Sep 30 – Oct 01, 2021)
TNW Conference is the most popular tech conference in Amsterdam, a 2-day technology event that brings together international technology executives, policymakers, top-tier investors, scale-ups, and startups for business and knowledge sharing. TNW Conference has grown into one of Europe’s leading technology events and the 15th edition of the conference will bring together 24.500 attendees from all over the world. Visitors include entrepreneurs, developers, marketing managers, CEOs and policymakers.
The Next Web Conference. Amsterdam, The Netherlands.
Skinive presented its solutions at the corporate stand, and also made a presentation on the stage of the Amsterdam Powerpod. Dozens of participants were interested in Skinive solutions. As a result of the conferences, Skinive found several warm contacts and continued negotiations on further cooperation.
Skinive Solutions at TNW Conference. October, 2021. The Next Web Conference Kirill Atstarov – Skinive CEO & Founder at The Next Web Conference
TNW Conference has been running since 2006 and has proven itself as the prime hub that gathers together technology agents driving business innovation.
TNW Conference
Thank you the Startup Amsterdam and Smart Health Amsterdam for the invitation and the opportunity to present Skinive solutions at TNW Conference!
2. World AI Summit (WSAI) (October 13-14, 2021)
WSAI is the world’s leading and internationally renowned AI summit gathering the global AI ecosystem of Enterprise, BigTech, Startups, Investors and Science, the brightest brains in AI as speakers every year in Amsterdam to tackle head-on the most burning AI issues and set the global AI agenda.
World AI Summit 2021. A community of 200,000 InspiredMinds from 167 countries worldwide! Skinive Solutions at World AI Summit
World AI Summit 2021 WSAI. All photos copyright by Skinive
3. Intelligent Health Conference 2021 (October 12, 2021)
Intelligent Health Conference is the world’s leading AI in Medicine Summit. Intelligent Health is the only large-scale, global summit series focused purely on AI in healthcare and bringing the global AI and health community together to advance discussions on how to apply AI and drive technological collaboration in healthcare.
Digital Health Meeting is a collaboration between DigitalHealthPartners & Smart Health Amsterdam, combining matchmaking and networking.
Digital Health Meeting. Amsterdam, The Netherlands. Digital Health Meeting. October 7, 2021. Kirill Atstarov – Skinive CEO & Founder at Digital health meeting
In addition, we are happy to announce that Skinive was selected as a finalist in 2 competitions: eAwards Netherlands 2021Final Top 10 and Blue Tulip Awards (Health&welness) Top 20!
The Blue Tulips Award programme and eAwards Netherlands are competitions that aim to connect innovators, build powerful collaborations and accelerate breakthrough innovations.
Our team is pleased to present a new application in the Skinive line – the Skinive Web Widget, based on our API service Skinive.Cloud .
We have a holistic dermatological artificial intelligence that can scan 30 skin conditions: all forms of skin cancer, viral diseases and acne.
Check the accuracy of our algorithm via the Skinive web widget below:
The Skinive Web Widget easily integrates into websites with any CMS-systems, from Tilda and WordPress to custom systems, and searches our image database for matches and returns the name of the skin condition.
Skinive will be able to increase user engagement and conversion in actions for the following categories of sites:
Skin Care Websites & Blogs
Health Information Sites
Telemedicine Platforms
Websites of clinics and beauty salons
Websites for healthcare professionals
Pharmaceutical companies specialized in dermatology
Patient Advocacy Sites
Do you want to install the Skinive web widget on your site?
We are proud to announce that Skinive received ISO-13485 Quality Management System (QMS *) Certificate (integrated with IEC 62304 and ISO 14971) for medical devices and software in December 2020.
Safety and quality are non-negotiable in medical devices, which is why our company’s management system is ISO 13485 certified. ISO 13485 standard is the progressive design, configuration, manufacture, installation and sale of medical devices that are safe for their intended use.
This success is the result of a continuous striving to improve our products and services and could not have been achieved without close collaboration within the company.
We are looking forward to new opportunities and projects with our old and new partners and clients!
These documents allowed us to obtain an official assessment of the risk class of our product from the European Competent Authority: I under the MDD Directive and IIa under the MDR Regulation. The implementation of ISO 13485 makes it possible to register our mobile application as a medical device in the European Union and receive the CE marking. In the process of developing the QMS, we have developed a user manual taking into account the requirements of the MHRA recommendation, MDR and mobile application labeling.
The ISO 13485 concept is applicable to companies that design, manufacture, and service medical devices. This standard specifies the requirements for a quality management system * that can be used by an organization in performing all stages of the life cycle of a medical device.
The process of implementing the Quality Management System is rather long and complicated, but it carries a number of advantages:
expansion of the sales market, since the Standard makes it possible to use our mobile application in all European countries, as well as in a number of other countries;
increasing the level of trust from customers;
increasing image and competitiveness;
the opportunity to participate in public procurement and obtain lucrative contracts.
We applied a clear system of work for each department of our company to allow ourselves to provide quality services to the consumer.
Artem Lian
CTO, Head of DataScience
“Considering that we develop software in the field of medicine, it is very important for us to build processes for the further development of the product in accordance with the regulatory requirements of the European Union and the United States. The QMS allows you to track all current regulatory documents through the Regulatory support process. “ – commented Artem Lyan, Skinive co-founder and CTO.
Kirill Atstarov
CEO, Co-founder
“The implementation of ISO 13485: 2016 is a key milestone on our journey towards full compliance with global health regulations. We look forward to supporting the global community of dermatologists and skin health professionals with Skinive” – commented Kirill Atstarov, Skinive CEO and Co-founder.
The Skinive quality review was carried out by a team of experts from XportKat, an international consulting company, the QMS audit is planned by a certification body authority based in Czech Republic, Prague.
To learn more about our solutions and benefits, schedule a video call with our experts https://calendly.com/skinive/
* QMS (quality management system) is a set of processes within an organization aimed at identifying, measuring, controlling, and improving core activities in accordance with customer requirements and regulatory requirements applicable to your organization’s medical devices and related services. Most countries have also recognized international standards published by committees such as the International Organization for Standardization (ISO) as a core requirement for compliance.
Note:This is not the current version of the report!
Skinive improved the accuracy of the neural network in 2021 and published the data. The latest version of the report is available at the link – https://shatiko.pp.ua/skinive-accuracy2021/
Table of contents
– ABSTRACT 1. INTRODUCTION 2. SKINIVE ALGORITHM TO ANALYZE THE SKIN LESION IMAGES
2.1 Nosologies & Classes
2.2 Neural network architecture
2.3 Data security 3. MATERIALS AND RESULTS
– CONCLUSION
– REFERENCE
– AI EXPERT REVIEW
– MEDICAL EXPERT REVIEW
Analysis of Skinive algorithm’s accuracy for risk assessment of skin conditions, based on machine learning algorithms.
Machine learning algorithms for medical imaging processing are now achieving expert accuracy and are being actively introduced into medical practice. However, there is no objective
assessment of the use of machine learning to classify skin lesions in a number of smartphone applications. The lack of objective methodologies and open data sets for evaluation of these
algorithms (as in case of e.g. Imagenet for general object recognition in images) hinders objective assessment by specialists and impedes the widespread use of this technology in public
health.
Objective
In this study, we experimentally evaluate the accuracy of Skinive algorithms and compare them with the previously published study on skin cancer risk assessment(1*).
Methods
This publication presents in detail the results of our smartphone application system. Skinive uses a machine-learning algorithm to calculate the skin pathologies’ risk rating. The algorithm
is trained on 63,955 images. All the images in the dataset have been assessed by dermatologists for risk.
To evaluate the sensitivity of the algorithm, 3 validation data sets are used:
(Pre) malignant – 285 cases of skin cancer and pre-cancerous conditions;
HPV – 285 cases of Human Papilloma Virus;
Acne – 285 cases of acne, milia, rosacea.
We calculate specificity on a separate set containing 6000 benign cases.
Results
To simulate an experiment, the authors prepared validation datasets with a similar distribution of the number of images by nosology and used the Skinive neural network to analyze the images and classify the levels of risk, similar to the example below:
* Sensitivity is defined as the ratio of the number of cases of skin pathology correctly determined by the algorithm (precancerous diseases and malignant tumors, acne and HPV) to the number of all clinically confirmed cases, respectively.
** Specificity is equal to the number of benign cases correctly classified by the algorithm as low risk (true negative cases), divided by the total number of all clinically confirmed benign cases.
The results obtained above followed closely the experimental setup proposed in 1 * in terms of relevant class distributions and a total number of cases.
Conclusions
The results of the accuracy of the neural network are comparable with the accuracy of dermatologists obtained in studies (5 *, 6 *) and can be considered as an expert system for supporting the adoption of a medical decision.
Pic: Health Professionals Accuracy Scale
The results of the comparative analysis cannot be interpreted unambiguously and can not be fully reliable, as in the datasets were used data from different sources. The lack of open data (photos) and common approach to validating decisions from different manufacturers do not allow independent benchmarking, needed to confirm the effectiveness of the method in general and objective comparison of existing solutions. Nevertheless, the results obtained above are on par with medical professionals and can be further improved with additional data and more optimized algorithms.
The lack of open validation datasets (images) and a common approach of different developers to decision validating do not allow them to conduct independent benchmarking, which is necessary to confirm the effectiveness of the method in general and objective comparison of existing solutions.
Future research is needed to define the role and assess the impact of mobile applications on the health system and its users and to further discuss the implementation of common methodologies to assess the effectiveness of mobile applications for assessing the risk of skin diseases.
P.S.
We are open to collaborative research with other datasientist teams and can provide access to our validated dataset upon request.
We are ready to provide a full version of our study for the following purposes:
publications in medical journals and print media;
reviews by medical and technical experts;
partners, corporate clients, investors (on NDA terms until the moment of official publication in open sources).
Note:This is not the current version of the report!
Skinive improved the accuracy of the neural network in 2021 and published the data. The latest version of the report is available at the link – https://shatiko.pp.ua/skinive-accuracy2021/
Research work described in this article was conducted by the authors during the Rockstart AI acceleration program in ‘s Hertogenbosch, the Netherlands . The purpose of the study was threefold:
Identify a representative and balanced set of images that can be made available to all researchers utilizing computer vision and machine learning for classification of skin lesions.
Create a benchmark for evaluation of all similar methods in this field.
Compare Skinive results to state of the art in the field on the basis of the defined benchmark and data; put those results in the context of medical professionals’ performance.
As part of the acceleration program, I have had weekly sessions with the Skinive technical team, overseeing the experimental setup, dataset composition, class distribution and other relevant aspects in creation of such a benchmark. To the best of my knowledge, the benchmark satisfies all the necessary criteria and is thus a good candidate for a golden standard for other researchers in the field in testing the performance of their algorithms.
In addition, the authors and I rigorously inspected results of all the experiments, including those shown in this report. The authors went to great lengths to ensure that these results are not only optimal for their use-case, but also comparable to state-of-the-art in the field as well as human medical experts.
On the article “Analysis of Skinive algorithm’s accuracy for risk assessment of skin conditions, based on machine learning algorithms.” (Authors: K.Atstarov, A.Lian, V.Shpudeiko, A.Ahushevich, I.Lichko).
This article is devoted to the urgent problem of dermatovenerology – the study of the prospects for the use in medicine of machine learning algorithms for processing medical images, which will improve the early diagnosis of skin oncopathologies. The aim of the study was to study the diagnostic accuracy of the Skinive mobile application and to compare the results with the previously published work of Skinvision B.V.
As a result of the studies, the authors established the sensitivity (79,6%-89.1%) and specificity (93.5%) of the Skinive mobile application, which indicates a high level of the diagnostic method, but there is a need to improve the sensitivity level for detecting skin cancer.
Conclusion: This article was written at a high scientific level. The structure of the article consistently reflects the logic of the study. It should be noted that the article is written in clear language, not overloaded with highly specialized terminology. The findings of the authors are well-founded. The results of the work may be useful to oncodermatologists, dermatovenerologists, as well as general practitioners.
Reference:
1. Accuracy of a smartphone application for triage of skin lesions based on machine learning algorithms
A. Udrea, G.D. Mitra, D. Costea, E.C. Noels, M. Wakkee, D.M. Siegel, JEADV; accepted for publication. T.M. de Carvalho, T.E.C. Nijsten. Published on September 08, 2019. https://onlinelibrary.wiley.com/doi/10.1111/jdv.15935
5. Assessing diagnostic skill in dermatology: a comparison between general practitioners and dermatologists.
Tran H1, Chen K, Lim AC, Jabbour J, Shumack S. Published in November, 2005
https://www.ncbi.nlm.nih.gov/pubmed/16197420
6. Comparison of dermatologic diagnoses by primary care practitioners and dermatologists. A review of the literature.
Federman DG1, Concato J, Kirsner RS. Published in April, 1999
https://www.ncbi.nlm.nih.gov/pubmed/10101989
7. The 2019 novel coronavirus disease (COVID-19) pandemic: A review of the current evidence.
Chatterjee P, Nagi N, Agarwal A, Das B, Banerjee S, Sarkar S, Gupta N,
Gangakhedkar RR. Published on March 30, 2020
https://www.ncbi.nlm.nih.gov/pubmed/32242874
After weeks of deliberation, Rockstart has finalized the selection and launched its 3rd AI accelerator program, which kicked off last Thursday night in ‘s-Hertogenbosch, the Netherlands. Attendees included mentors, investors as well as other start-ups. These talented entrepreneurs from ten startups across eight different countries will spend the next six months building and refining their products, talking to mentors, and connecting with both potential customers and investors. The first AI program is already showing success with three alumni raising a total of €1.2 million within the first six months after the end of the program, not to mention the vast range of technology collaborations between startups and corporates that have spiraled out from the program.
“We’re excited to welcome startups from all over the world to this third edition of the AI program. Once again, we have a diverse cohort of technology startups utilizing datasets in industries such as healthcare, education, design, database infrastructure, transportation, trading (fiat and crypto), and graphic design – just to name a few,” said Rune Theill, CEO at Rockstart.
As a part of the Rockstart accelerator, startups also gain access to start-of-the-art research, know-how, and talent through Rockstart’s partnership with Jheronimus Academy of Data Science (JADS). JADS is a Master’s and Ph.D. Data Science University focused on intertwining data science with entrepreneurship.
Currently, AI partners, including in-kind, consist of; AWS, Braincreators, Google, IBM Benvalor, Cisco and more. Apart from this strong lineup, startups also have access to Rockstart’s network of experienced mentors, investors, and experts. These mentors will support and guide the startups chosen for this year’s program through the challenges of technology and business development. “I am thrilled to see the amount and quality of startups who applied for this 3rd accelerator program. Even more excited to announce the ten finalists today. I am confident that each of them will bring a real impact in their respective industries,” said Raymond Alves, AI Program Director at Rockstart.
Doctor Roitberg’s Doctor Smart online medical consultation service, in partnership with the Skinive AI project, has launched a new service – skin scoring based on artificial intelligence (AI). Users can scan their moles (pigmented nevus) in a few seconds by uploading a photo to the application. The neural network will analyze the image and prepare a conclusion whether the formation is dangerous.
Now, users of the Doctor Smart online clinic can now scan their moles (pigmented nevus) using Skinive technology in a few seconds by uploading a photo to the application. The neural network will analyze the image and prepare a conclusion whether the formation is dangerous.
The creation of a health assistant based on neural networks Doctor Smart aims to reduce the burden on doctors and make quality medical services more accessible for users, explained Doctor Smart co-founder Pavel Roitberg:
“We aim to create a multidisciplinary health assistant based on artificial intelligence. Our team, together with partners, develops neural network solutions in the field of medicine, in the future such technologies will help to significantly increase the accuracy and speed of primary diagnostics and reduce operating costs. We see the need of users for high-quality and affordable remote service and plan to launch new products. “
Doctor Smart Team
Recommendations for Skinive AI skin health scoring tests: ● You have a lot of moles on your body, papillomas, genital warts, age spots; ● One of the relatives was ill with melanoma – a high predisposition to skin cancer; ● You spend a lot of time in direct sunlight; ● You noticed an increase in the number of moles, an expansion of the boundaries of age spots.
Skinive.Cloud is a development by WiseAI llc, which determines the pathology of pigmented lesions from images (photo). This is the second product based on medical AI as part of the service. Earlier, Doctor Smart integrated the Care Mentor AI neural network for radiological image analysis. The neural network allows the patient to analyze the study even before the doctor’s appointment or to get a second opinion on the diagnosis already made.
The skin scoring service is already available in the Doctor Smart service – it has no contraindications and you only need a smartphone to receive it. Until the end of August, you can use it for free.
We are pleased to announce that Skinive is now available on Microsoft Azure Marketplace and we became a part of Microsoft Partner Network!
The Skinive team is open to collaborating with third-party solution developers at DigitalHealth and has announced the publication of its solution in the Azure cloud!
Skinive.Cloud is available to third-party application developers to integrate with web applications, mobile applications and medical systems to improve the quality of skin care users.
The Skinive API service is designed to reduce the problem of the availability of mobile screening tests for end users, as well as for medical professionals: skin care specialists, cosmetologists, general practitioners, and dermatologists.
While you most likely don’t realize it, machine learning is often used in your daily life. For example, when social media suggests tagging your friends in pictures because it recognizes them, or the spam filter on your email account removing unwanted emails. In healthcare, machine learning also takes its part in recognizing skin cancer. Machine learning has been used in hospitals for many years, but now you can use it yourself to track your health in the comfort of your home!
Over the past years, Skinive has made great progress towards developing an application that is significantly reliable in recognizing dangerous skin lesions. Skinive started off with a ‘rule-based’ system which went through every picture and checked skin lesions for certain characteristics to determine risk. Even though this algorithm has helped us detect the risk of thousands of dangerous lesions, we are continuously looking to improve its accuracy.
That is why we are very excited to announce that in August 2018 we introduced a smart system that detects dangerous skin lesions completely based on a machine learning algorithm.
What is Machine Learning?
Here is a great animation that explains it in two minutes.
How is Machine Learning used in the Skinive application?
We have trained the Skinive algorithm with large quantities of images which were previously assessed by our team of dermatologists. The algorithm learns which lesions are dangerous and which ones are not. We continuously train and improve our algorithm with new sets of images. From now on, all the pictures submitted through the Skinive application go through this algorithm.
It is common for doctors to ask a second opinion, and so at this moment, every photo is also reviewed by our in-house dermatologists and image recognition experts. We have set up this process to assist the algorithm to become more accurate and to make sure that our dermatologists agree with the risk indication. The best part, however, is that we are training our algorithm to become on a par with the best dermatologists.
Every photo of skin spots makes our algorithm smarter at detecting skin cancer risk. Contribute to our mission of saving lives by using Skinive . Try it now!
