Biosensor for personalized drug monitoring of antibiotics

The right antibiotic dosage can be crucial for effective treatment. Currently, determining the optimal individual dose for a patient requires analyzing blood samples in a lab—a process that is laborious, expensive, and typically takes a day to produce results. As part of her doctoral research, Dr. H. Ceren Ates developed and tested a biosensor that can measure antibiotic concentrations in various body fluids and deliver results within a short time. This opens up new possibilities for personalized therapies.

The sensor detects concentrations of beta-lactam antibiotics - a widely used class that includes penicillin. It enables non-invasive concentration measurements via exhaled breath condensate - a novel capability - while also supporting established blood sample analysis, with results available in 30 minutes. Additionally, it can measure concentrations in other bodily fluids, such as saliva or urine.

Compact, fast, and accurate

“Accuracy, speed, and ease of use are key when monitoring antibiotic concentrations,” says H. Ceren Ates. The new sensor can analyze up to eight different samples or sample types at once—for example, blood and breath condensate—making it a multiplexed sensor. It is compact and can be used directly at the patient’s bedside or in a doctor’s office, eliminating the need for complex laboratory procedures. In animal studies, the sensor performed as well as standard lab procedures using blood samples. Since the breath-based method is non-invasive and fast, it is particularly suited for continuous monitoring.

The biosensor requires only tiny amounts of fluid - less than one microliter, just a fraction of a drop of water. It uses an antibody-free testing method where the antibiotic concentration is reflected by the electrical current - the lower the current, the higher the antibiotic concentration.

Personalized antibiotic treatment

“The effective antibiotics dose can vary significantly from person to person,” says H. Ceren Ates. With this in mind, she developed an AI-based method that uses real-time patient data to monitor and predict individual patient recovery trajectories. The aim is to tailor the dose based on each patient’s unique recovery profile. Like a GPS, the model can serve as a guide for doctors: it analyzes a patient's current condition and evaluates the effectiveness of the current dosage strategy based on previous measurements, providing quantitative information about how quickly the patient is approaching or moving away from a healthy state. Based on this, doctors can adjust the dosage individually. First tests with the system were successful, and it will now be evaluated in further studies.

Bertha Benz Prize and Gips-Schüle Young Talent Award

For her dissertation “Multiplexed biosensors toward smart therapeutic drug management of antibiotics” H. Ceren Ates, postdoctoral researcher in the group of Prof. Can Dincer, was recently honored with two awards: the 2025 Bertha Benz Prize from the Daimler and Benz Foundation and the 2025 Gips-Schüle Young Talent Award from the Gips-Schüle Foundation.

“I’m incredibly honored to receive these awards,” says H. Ceren Ates. “In the long term, we aim to develop the system further so that it can be integrated into wearables. We also plan to expand the system to detect inflammation markers, among other things.”