Descrizione del progetto
Un cervello miniaturizzato a base di «sinapsi» organiche classifica le cellule in un dispositivo da tavolo
I computer sono stati a lungo equiparati al cervello, ma solo ora ci stiamo avvicinando alle corrette elaborazioni e architetture neuromorfiche. L’intelligenza artificiale ha cominciato con una logica tradizionale basata su regole ed è progredita verso l’apprendimento basato sull’inferenza e sull’esperienza. Per raggiungere il livello successivo in cui le decisioni possono essere elaborate in situazioni nuove che non sono in alcun modo predeterminate, alcuni scienziati si stanno orientando verso architetture persino più antropomorfe. I ricercatori dietro il progetto BIOMORPHIC, finanziato dall’UE, hanno sviluppato in precedenza una sinapsi artificiale organica in grado di replicare la giunzione tramite la quale due neuroni trasmettono le informazioni sotto forma di segnale elettrochimico. BIOMORPHIC ha intenzione di innalzare incredibilmente il livello, creando una rete interconnessa di tali «sinapsi» in grado di classificare le cellule in un dispositivo microfluidico economico.
Obiettivo
Brain-inspired (neuromorphic) computing has recently demonstrated advancements in pattern and image recognition as well as classification of unstructured (big) data. However, the volatility and energy required for neuromorphic devices presented to date significantly complicate the path to achieve the interconnectivity and efficiency of the brain. In previous work, recently published in Nature Materials, the PI has demonstrated a low-cost solution to these drawbacks: an organic artificial synapse as a building-block for organic neuromorphics. The conductance of this single synapse can be accurately tuned by controlled ion injection in the conductive polymer, which could trigger unprecedented low-energy analogue computing.
Hence, the major challenge in the largely unexplored field of organic neuromorphics, is to create an interconnected network of these synapses to obtain a true neuromorphic array which will not only be exceptionally pioneering in materials research for neuromorphics and machine-learning, but can also be adopted in a multitude of vital medical research devices. BIOMORPHIC will develop a unique brain-inspired organic lab-on-a-chip in which microfluidics integrated with sensors, collecting characteristics of biological cells, will serve as input to the neuromorphic array. BIOMORPHIC will combine modular microfluidics and machine-learning to develop a novel platform for low-cost lab-on-a-chip devices capable of on-chip cell classification.
In particular, BIOMORPHIC will focus on the detection of circulating tumour cells (CTC). Current methods for the detection of cancer are generally invasive, whereas analysing CTCs in blood offers a highly desired alternative. However, accurately detecting and isolating these cells remains a challenge due to their low prevalence and large variability. The strength of neuromorphics precisely lies in finding patterns in such variable data, which will result in a ground-breaking CTC classification lab-on-a-chip.
Campo scientifico
- natural sciencesphysical sciencesclassical mechanicsfluid mechanicsmicrofluidics
- engineering and technologyother engineering and technologiesmicrotechnologylab on a chip
- natural scienceschemical sciencespolymer sciences
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringsensors
- medical and health sciencesclinical medicineoncology
Programma(i)
Argomento(i)
Meccanismo di finanziamento
ERC-STG - Starting GrantIstituzione ospitante
5612 AE Eindhoven
Paesi Bassi