In-Memory Sensing (In-Me Project)

Project Title: “In-Me: In-Memory Sensing ”

Project Duration: 02/2022 – 02/2025

Project Description:

In the modern era, artificial intelligence and the Internet of Things have enabled us to develop smart edge devices that can sense, think, and communicate. These devices are based on Von Neumann's architecture which greatly contributed to the advancement of computing technologies. In spite of this, one of the drawbacks of Von Neumann's architecture is that the continuous transfer of data between memory and processors requires a considerable amount of processing time and power, and thereby represents a bottleneck in the process. Therefore, this project has as its objective to develop a novel architecture allowing the development of the new concept of "in-memory sensing" with the capability of sensing, computing, and memorizing data in the edge. As a proof of concept, prostate cancer markers will be selected as the target analyte for the in-memory sensor.

Even if the measurement of a biomarker in the blood of a patient could be indicative of cancer progression, it is well known that the percentage of false positives and false negatives are still far too high to obtain a definitive diagnosis. As an example, the so-called Prostate-Specific Antigen (PSA), a serine protease involved in the liquefaction of seminal fluid, is still considered the best biomarker for prostate cancer (indicating also prostate dysfunction). To minimize the possibility of false-positive results on PSA detection, both arithmetic and logic operations must be utilized. Accordingly, In-Me aims to develop multi-memristor sensory platforms that will detect the concentration of biomarkers (e.g., PSA) in blood samples, while simultaneously calculating and memorizing the required cancer risk scores.

The general objective of the project is to develop an array of flexible and printed sensors and biosensors that form a memristive architecture. The research activity will focus on the fabrication and characterization of the components in function of the active material and substrate use, and the realization of the final circuits. In particular, the following activities are foreseen:

• Simulations based on equivalent circuits will be performed to demonstrate “in memory sensing and computing” by design,
• Preparation of inks based on carbon nanotubes and polymeric materials,
• Realization of physical and chemical memristive-based sensors and biosensors and their characterization, and
• Integration of these sensors and biosensors to realize fully memristive systems in biomedical environments.

In-Me is an interdisciplinary project with a strong collaboration between the Free University of Bozen-Bolzano, and the Swiss partners the research institution in Lausanne - EPFL (Swiss Federal Institute of Technology Lausanne), and CHUV hospital in Lausanne (the Centre Hospitalier Universitaire Vaudois).

Principal Investigator: Prof. Paolo Lugli

Project Collaborators:
Prof. Sandro Carrara , EPFL, Lausanne, Switzerland
Dr/PhD. Rudolph Dominik Berthold , CHUV, Lausanne, Switzerland