Feature stabilization in convolutional neural networks using Proportional Integral Controller for lung nodule classification

V. Sangeetha, S. Kalaivani · 2026

Read Feature stabilization in convolutional neural networks using Proportional Integral Controller for lung nodule classification by V. Sangeetha, S. Kalaivani (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13171795.txt.

Description

Sangeetha and Kalaivani steady the inner features of a neural network that sorts lung nodules on CT scans. Their study appeared in twenty twenty-six. Shifting brightness, noise, and unstable features during training make this sorting task unreliable, above all with scarce clinical data.

Their design borrows a classic control tool, the proportional-integral controller, to correct feature drift as learning proceeds. The controller is placed before the network or woven inside it, with settings tuned by hand or learned from data. One part reacts to instant errors; the other repays accumulated ones. On a lung cancer image set, the method reached ninety-six percent accuracy with no false positives. Checks confirmed the gain.

Catalog record

CREATORV. Sangeetha, S. Kalaivani
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE89 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13171795

Browse the full library · About the Tower of Babel