Artificial intelligence (AI) predicts the best treatment for patients with melanoma. The AI model was developed as part of the Data Science Experience (DSX) project, which was set up by the Isala Hospital.
“My desire is to improve patient care by leveraging available data,” said Richard Brohet, genetic epidemiologist, statistician and data scientist at the Isala Academy. Brohet started the DSX project together with colleagues.
“In the first instance, I want to do this for oncology patients. DSX uses all medical data in Isala, or Big Data. We discussed with doctors what kind of algorithm with AI (artificial intelligence) we wanted to develop first. The choice fell on melanoma. That is why I started this pilot study together with internist-oncologist Jan Willem de Groot.”
What is the best treatment?
Patients with cancer are faced with various issues. Why did it arise and what is the best treatment? “For oncology patients, these questions are difficult to answer. Cancer can arise from various causes, such as lifestyle, hormones, environment, genetic abnormalities, heredity, or a combination of these factors. Also, cancer treatment depends on personal, clinical and genetic characteristics. Because every tumor and therefore every patient is unique, we want a tailor-made treatment so that survival is better and the chance of a treatment that does not work is smaller. In addition to chemotherapy, patients with melanoma now also receive targeted therapy or immunotherapy. This works for some patients, but not for others. How is that possible? AI could help tremendously in answering this question and delivering tailor-made treatment. That is what we are now trying to do with DSX”, explains Brohet.
The AI model developed as part of the DSX project uses structured and unstructured data for analysis. For structured data, consider gender, age, weight and size of the tumor. Unstructured data includes data entered in open text fields. This includes how someone is doing, their lifestyle and profession, and how the patient is coping with the disease. This specifically concerns information that does not fit in a checklist, but is important.
To see conections
De Groot: “By also using that data, you can search for what you don’t expect and see connections. It also gives us insight into the outcomes of the various treatments over time. In this way we can capture and analyze more complex data better and better and we can better see what predictive factors are.”
The basic AI model that has now been developed predicts the best treatment for a patient with a melanoma. However, the model still needs further validation. De Groot reports that if he enters the data of a known patient, the model indeed predicts the correct treatment. The algorithm now needs to grow, which the makers want to do by adding more patients, including from other melanoma centers. For example, they want to further optimize the predictions and thus offer patients tailor-made solutions. De Groot: “People are often looking for medicines that can prolong life by a few months. My opinion is that we can invest better in the quality of life. We can do this by learning from the available data in order to decide together with the patient about the most appropriate treatment.”
Determining Lung Cancer Risk
For some time now, researchers have been working on AI models that support the treatment of oncology patients. This is how a . judges AI-algorithm developed by researchers at the Radboudumc lung nodules. The model predicts the risk of these nodules developing into tumors. Detecting tumors at an early stage increases the chance of successful treatment and reduces the risk of death from lung cancer. However, many of the small nodules that show up on CT scans are benign. These can be left alone, while malignant nodules need to be addressed. The algorithm distinguishes between benign and malignant lumps.
A second example is a algorithm developed by researchers from Eindhoven University of Technology. This is a machine learning model that predicts whether immunotherapy will work effectively in a patient with cancer. Among other things, our immune system works to eliminate threats in our body, including cancer. However, cancer cells can disable immune cells, preventing their destruction. Immunotherapy can reverse this process in some cases. The algorithm predicts the chance of successful treatment with immunotherapy.
Fighting uterine cancer
Through the PRESCRIP-TEC research project, which includes the LUMC involved, researchers are working on AI that will help fight uterine cancer. This cancer is caused by the human papilloma virus (HPV). Women with chronic HPV infection develop precancerous cervical cancer. These can be traced through screenings. In the Netherlands, screenings via smears are commonplace. However, in many low- and middle-income countries, access to health care is often poor and screenings are rare.
A self-test can provide a solution, after which a relatively simple treatment is possible. However, these are relatively labour-intensive in remote areas, while rapid intervention is important. Through the PRESCRIP-TEC project, the researchers are working on AI that helps assess self-tests that women perform on themselves with a cotton swab. If they have HPV and need to be screened, AI technology will help health professionals recognize the discoloration of the cervix.
Author: Wouter Hoeffnagel
Foto: Gordon Johnson via Pixabay
–
