Machine Learning Assisted Sensitivity of Triple-Material Gate -Stack Gate-All-Around (TM-GS-GAA) MOSFET based Biosensor

In this paper, the variation in dielectric of TM-GS-GAA MOSFET device has been considered for the detection of biomolecule. The gate stack has high and low– K dielectric constant. The surface potential and drain current are analysed for high-k gate dielectric materials like TiO2, ZrO2 and Al2O3. The threshold voltage variation and the Ion/Ioff ratio serve as the primary measures for determining sensitivity. Furthermore, the machine learning based decision tree algorithm has been proposed to investigate the sensitivity of biomolecules with variation in nanogap cavity. The mean squared error (mse) and R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>-score has been calculated for the proposed algorithm for the explored device. It shows mse as 0.02 and 99 % as R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>-score which indicates the proposed algorithm achieved the same accuracy as that of device simulation. The TM-GS-GAA MOSFET based biosensor has been compared with DM-JL MOSFET and shows an excellent enhancement in the sensitivity. The new device is designed for advance biosensing applications, offering innovative solutions for detecting and analysing biological materials.