Evidence 01
- More people are diagnosed with skin cancer each year in the U.S. than all other cancers combined. Skin cancer is the fourth most common cancer worldwide.(1)
01 / Detection
Computer vision and Deep learning
AI diagnosticsof skin cancer
Decision support FDA 510k exempt
Enter the analysis
02 / Clinical problem
The problem
Evidence 02
- Melanoma is the most aggressive skin cancer. For patients with early stage and localized melanoma the survival rate is 99%; yet when melanoma has metastasized it is more resistant to current therapies and the survival rate drops to 15-20%. Progression from early to metastatic stages can occur in months, making the prompt diagnosis of melanoma particularly important.(2)
Evidence 03
- Tissue biopsy is the only definite diagnostic test for skin cancer. With the increasing demand for quality healthcare globally and increased health screening, the number of pathology samples to be analyzed is also increasing. Pathology experts are required to review non-cancerous specimens in more than 90% of cases.
- The number of practicing pathologists retiring coupled with the years of practice required to achieve an expert level of competency is causing a drastically increasing shortage of expert pathologists worldwide.(8) Patient care is in jeopardy because specimen volume has risen by more than 40% while the number of professionals in the pathology field has declined.
Clinical evidence
Facts and sources
Review detailed facts and citations
Additional context
- Delayed responses and the high cost of pathologist opinions result in reduced time to implement proper care at the early stages of disease when the tumor is likely to respond to treatment.
- Solutions helping to instantaneously differentiate benign neoplasms from potentially cancerous lesions can radically increase throughput, shorten response times and reduce costs.
- Access to expert pathology opinions is limited in low-income countries, with less than 30% having access to pathology services in the public sector as compared to more than 90% in high-income countries.(1) Applying digital pathology and AI algorithms to cancer diagnostics is the only viable option to improve access.
- According to studies in the U.S., pathologists disagree on a diagnose in 20-25% of cases on average, particularly in early stages when the case is not clear-cut, which generates demand for second opinions and expert consultations worldwide.(4)
- Diagnostic errors contribute to approximately 10% of patient deaths and are the leading cause of medical malpractice payouts.(5)
Key facts
- Skin cancer is the most common cancer type in the U.S. with more than 500 thousand cases estimated in 2018 (including non-melanoma skin cancer), which is two-times higher than breast cancer, the second most common cancer. One in five Americans will develop skin cancer by the age of 70 and represents the fourth most common cancer type worldwide, with an incidence of 1.3 million. The incidence of various skin cancer types grows at 2-9% annually. (1),(6)
- For very early-stage melanoma limited to the site of origin, the 5-year survival rate is 99%. Survival for melanoma invading nearby lymph nodes is 63%. If melanoma has metastasized, survival rate is 15-20%.(2)
- 5% of skin tissue biopsies in the U.S. are diagnosed as melanoma.(3)
- According to a study in 10 U.S. states conducted by Elmore, the discordance rates between pathologists assessing skin biopsies ranged from 57% to 75% in middle-ground classes, including disagreements among themselves ranging from 37% to 65%, resulting in incorrect diagnoses in 20-25 cases.(4)
- The number of active pathologists in the US decreased by 18% and the diagnostic workload per pathologist rose by 42% in the last 10 years, 63% of active pathologists are age 55 or older(8)
- The annual cost of treating skin cancers in the U.S. is estimated at $8.1 billion in 2000s, with approximately $4.8 billion for non-melanoma skin cancers and $3.3 billion for melanoma.(7)
03 / WSI analysis
AI image recognition for skin cancer diagnostics
Analysis 01
AI-based classification of skin tumor sub types
A supervised neural network learns from example whole-slide histopathology (H&E) images (WSI) annotated by top dermatopathology experts. The trained model is then used to return an accurate, robust and fast histopathology tumor type and sub type. The solution doesn’t require additional stains, patient or clinical information.
Analysis 02
Scanner and image quality agnostic
The technology works with images produced by all modern scanning equipment and is able to overcome staining and histology variability of original glass slides such as over-staining, tissue folds, bubbles, dust, ink markings, scanning quality variability.
Analysis 03
High accuracy on all common tumor types
Our network is capable of recognizing 40 different skin tumor morphological sub types. Over 20,000 images from 3500 cases annotated by top dermatopathology experts were used in the training. Testing on a set of 6700 images demonstrated an accuracy of over 98%.
Analysis 04
Accessible in a secure cloud
Our standalone solution is accessible in a secure cloud, can be integrated with Digital Pathology and Workflow systems.
04 / Decision support
Analysisbecomes clarity.
Our standalone solution is accessible in a secure cloud, can be integrated with Digital Pathology and Workflow systems.
05 / Performance
Evidence at scale
98%
% accuracy
01 / 04
40
Recognized diagnoses
02 / 04
6700
Tested images
03 / 04
10
Seconds per WSI
04 / 04
06 / Benefits
How our solution can help you
01 / 05
Healthcare providers
Outsourcing of pathology interpretation to the most suitable pathologist or lab based on their expertise, pricing, and availability; applicable to 2nd opinions as well. Improve accuracy and turnaround times, reduce the cost of incorrect treatment.
02 / 05
Anatomic pathology laboratories
Enhance the accuracy and reduce the costs of diagnosis by
Detailed applications
- Smart triage: distribute cases among pathologists with various levels of expertise to balance workloads, unburden senior pathologists from reviewing simple cases and thus remove bottlenecks, involve general pathologists in reviewing simple specialized cases, enable efficient outsourcing and remote work to increase lab capacity, support M&A and optimize costs.
- Prioritization: prioritize caseload for each pathologist to focus on urgent malignant cases first, improve turnaround times where it matters most, reduce the risk of late diagnosis.
- 2nd reading: additional opinion on diagnosis to increase the confidence of a pathologist in the decision, detection of the tumor-affected tissue on WSI, calculation of tumor depth and margins, and drawing attention to malignant and pre-malignant areas if they were overlooked.
- Quality Control: automatically spot all situations of disagreement between the algorithm and a pathologist and submit them to a quality assurance specialist.
03 / 05
Digital Pathology Software providers
Integration with diagnostic workflow to enable triage, prioritization, 2nd reading, and quality control on dermatopathology caseload to increase workflow efficiency, improve accuracy and speed. Fast and quantitative evaluation of images in the cloud. Detection of the tumor-affected tissue on WSI, calculation of tumor depth and margins.
04 / 05
Insurance Plans
While deployed at pathology labs, Mechanomind solution helps to reduce or eliminate the potential for misdiagnosis of biopsies, reduce the cost of diagnostics, response times, and subsequent healthcare expenses, improve treatment decisions and patient care.
05 / 05
Research and educational organizations
Our cloud solution can be used in research, allowing remote pathology review and undergraduate teaching of histology, residency/trainee pathology training and professional training in specialist areas of pathology.
Contact