Hennigsdorf/Berlin, September 25, 2019 – To diagnose an acute appendicitis correctly often remains a challenge. Especially in young patients the interpretation of symptoms is often difficult. In cooperation with Charité – University Medicine Berlin, OakLabs GmbH has now developed a model for a more reliable and differentiated diagnosis. In the future, it could make it easier for doctors to decide whether surgery is necessary or not.
Acute appendicitis remains one of the most common reasons for emergency surgery in children and adolescents. In industrialized countries, the lifelong risk of this is 7-9%. An important question that doctors are confronted with is whether it is a complicated or uncomplicated inflammation. While complicated inflammations require surgical removal of the vermiform appendix (appendectomy), current studies suggest that treatment with antibiotics may be sufficient in uncomplicated cases.
In September 2019, scientists from the Charité and a team of physicists and biochemists from OakLabs published a pioneering study in the renowned professional journal PLOS ONE. The research group retrospectively developed biomarker signatures based on standard blood and inflammation values as well as simple ultrasound parameters using algorithms from artificial intelligence. These allow the diagnosis of appendicitis on the one hand and the differentiation between uncomplicated and complicated cases on the other.
In contrast to conventional diagnostic methods, this method is independent of expert interpretation and significantly exceeds all previous methods in terms of accuracy.
The new method was able to correctly diagnose 93% of young appendicitis patients in retrospect. In addition, the use of the method could identify those patients whose symptoms were not caused by appendicitis. Thus, two out of three young people with suspected appendicitis could be saved from unnecessary surgery. The biomarker signatures also diagnosed 95% of complicated appendicitis cases and every third case of uncomplicated appendictis. In this way, the test could enable antibiotic instead of surgical treatment in one in three patients with uncomplicated appendicitis in the future.
"We are very pleased that our method has such a high expressiveness. Our goal is to further develop the method now in order to save many of our young patients unnecessary surgery in the future," says Dr. Marc Reismann, head of the study at the Charité Clinic for Pediatric Surgery.
"Based on standard laboratory values, the AI was able to significantly improve current diagnostics," adds Dr. Martina Schad, CEO of OakLabs GmbH. "The next step is to use more complex molecular data to further improve the differentiation between complicated and uncomplicated appendicitis."
The scientific publication of the results at PLOS ONE can be found at
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0222030.
OakLabs GmbH
Katrin Wahl
Head of Marketing & Communications
Neuendorfstr. 16b
16761 Hennigsdorf / Berlin
Germany
https://www.oak-labs.com
wahl@oak-labs.com
Phone: +49 3302 207 13 93
OakLabs is a global leader in biomarker development for diagnostics and precision medicine. Since its foundation in 2011, the interdisciplinary team of quantum physicists and (bio)chemists has been working together with international customers from the fields of pharmaceuticals, diagnostics and health care. The most important building block for success is exact algorithms based on artificial intelligence.
OakLabs is a German company with headquarters in Hennigsdorf near Berlin and a subsidiary in Cambridge, USA. For more information, please visit our website (https://www.oak-labs.com).
