Machine Learning Improves the Diagnosis of Patients with Head and Neck Cancers

Researchers from Charité - Universitätsmedizin Berlin and the German Cancer Consortium (DKTK) have successfully solved a longstanding problem in the diagnosis of head and neck cancers. Working alongside colleagues from Technische Universität (TU) Berlin, the researchers used artificial intelligence to develop a new classification method which identifies the primary origins of cancerous tissue based on chemical DNA changes. The potential for introduction into routine medical practice is currently being tested. Results from this research have been published in Science Translational Medicine.

Every year, more than 17,000 people in Germany are diagnosed with head and neck cancers. These include cancers of the oral cavity, larynx and nose, but can also affect other areas of the head and neck. Some head and neck cancer patients will also develop lung cancer. "In the large majority of cases, it is impossible to determine whether these represent pulmonary metastases of the patient's head and neck cancer or a second primary cancer, i.e. primary lung cancer," explains Prof. Dr. Frederick Klauschen of Charité's Institute of Pathology, who co-led the study alongside Prof. Dr. David Capper of Charité's Department of Neuropathology. "This distinction is hugely important in the treatment of people affected by these cancers," emphasizes Prof. Klauschen, adding: "While surgery may provide a cure in patients with localized lung cancers, patients with metastatic head and neck cancers fare significantly worse in terms of survival and will require treatments such as chemoradiotherapy."

When trying to distinguish between metastases and a second primary tumor, pathologists will usually use established techniques such as analyzing the cancer's microstructure and detecting characteristic proteins in the tissue. However, due to the marked similarities between head and neck cancers and lung cancers in this regard, these tests are usually inconclusive. "In order to solve this problem, we tested tissue samples for a specific chemical alteration known as DNA methylation," explains Prof. Capper who, like Prof. Klauschen, is a Scientific Member of the DKTK in Berlin. He adds: "We know from earlier studies that DNA methylation patterns in cancer cells are highly dependent on the organ in which the cancer originated."

Working with Prof. Dr. Klaus-Robert Müller, Professor for Machine Learning at TU Berlin, the research group employed artificial intelligence-based methods to render this information useful in practice. The researchers used DNA methylation data from several hundred head and neck and lung cancers in order to train a deep neural network to distinguish between the two types of cancer. "Our neural network is now able to distinguish between lung cancers and head and neck cancer metastases in the majority of cases, achieving an accuracy of over 99 percent," emphasizes Prof. Klauschen. He continues: "To ensure that patients with head and neck cancers and additional lung cancers will benefit from the results of our study as quickly as possible, we are currently in the process of testing the implementation of this diagnostic method in routine practice. This will include a prospective validation study to ensure that the new method can be made available to all affected patients."

Having worked alongside the researchers from Charité, the Director of the Berlin Center for Machine Learning (BZML), Prof. Müller, is similarly delighted at their results: "Artificial intelligence is playing an increasingly important role, not only in our daily lives and in industry, but also in natural sciences and medical research. The use of artificial intelligence is, however, particularly complex within the medical field; this is why, until now, research findings have only rarely delivered direct benefits for patients. This could now be about to change."

Philipp Jurmeister, Michael Bockmayr, Philipp Seegerer, Teresa Bockmayr, Denise Treue, Grégoire Montavon, Claudia Vollbrecht, Alexander Arnold, Daniel Teichmann, Keno Bressem, Ulrich Schüller, Maximilian von Laffert, Klaus-Robert Müller, David Capper, Frederick Klauschen.
Machine learning analysis of DNA methylation profiles distinguishes primary lung squamous cell carcinomas from head and neck metastases.
Science Translational Medicine, 11 Sep 2019: Vol. 11, Issue 509. doi: 10.1126/scitranslmed.aaw8513

Most Popular Now

Apple Health Records Available for Allsc…

Allscripts (NASDAQ: MDRX) announced that Apple Health Records is now available for Allscripts Sunrise™, TouchWorks® and Professional EHR™ clients and their patients. Health Records brings together hospitals, clinics and the...

Philips Signs Agreement to Create Taiwan…

Royal Philips (NYSE: PHG, AEX: PHIA), a global leader in health technology, today announced that Taipei Veterans General Hospital (TPVGH) will utilize the Philips IntelliSite Pathology Solution to transform its...

Machine Learning Improves the Diagnosis …

Researchers from Charité - Universitätsmedizin Berlin and the German Cancer Consortium (DKTK) have successfully solved a longstanding problem in the diagnosis of head and neck cancers. Working alongside colleagues from...

Robotic Thread is Designed to Slip throu…

MIT engineers have developed a magnetically steerable, thread-like robot that can actively glide through narrow, winding pathways, such as the labrynthine vasculature of the brain. In the future, this robotic...

St Helens and Knowsley Advance with Ambi…

St Helens and Knowsley Teaching Hospitals NHS Trusthas successfully gone live with System C’s CareFlow Vitals as part of its ambitious strategy to accelerate digitisation and become a digital exemplar...

The Future of Mind Control

Electrodes implanted in the brain help alleviate symptoms like the intrusive tremors associated with Parkinson's disease. But current probes face limitations due to their size and inflexibility. "The brain is...

Experimental Validation Confirms the Abi…

Insilico Medicine, a global leader in artificial intelligence for drug discovery, today announced the publication of a paper titled, "Deep learning enables rapid identification of potent DDR1 kinase inhibitors," in...

Using a Smartphone to Detect Norovirus

A little bit of norovirus - the highly infectious microbe that causes about 20 million cases of food poisoning in the United States each year - goes a long way...

Medical Informatics Europe Conference 20…

28 April - 1 May 2020, Geneva, Switzerland. The European Federation of Medical Informatics (EFMI) presents the 30th Medical Informatics Europe conference (MIE) at the Geneva International Conference Center (CICG). This...

Computer Model could Help Test New Sickl…

A team of Brown University researchers has developed a new computer model that simulates the way red blood cells become misshapen by sickle cell disease. The model, described in a...