In Digital Image Processing (DIP), image enhancement using univalent functions entails mapping pixel intensity values (gray levels) using one-to-one, invertible complex functions, frequently in the frequency domain or through specialized transformations, to stretch, compress, or redistribute intensities for improved visual quality or feature detection, such as boosting contrast or lowering noise, by making sure each input pixel maps uniquely to an output pixel, going beyond simple linear/non-linear transforms. In this paper, for the new subclasses \({\textbf{M}}_{\nu ,\wp }^{\varpi }(\varrho , \vartheta )\) and \({\textbf{G}}_{\nu ,\wp }^{\varpi }(\varrho , \vartheta )\) of \({\mathcal {A}}\) in association with normalized form of Miller-Ross-type functions explored and the key coefficient inequality as given below \(\begin{aligned} |a_n|\le \frac{\vartheta |\varpi |}{\{[1+\varrho (n-1)](\vartheta |\varpi |-1)+n\} \Lambda _n} \end{aligned}\) where \(\zeta \in {\mathbb {V}} = \{\zeta \in {\mathbb {C}},\,\, |\zeta |<1 \},\,\, \varpi \in {\mathbb {C}}-\{0\}, \,\, 0<\vartheta \le 1,\) \(\varrho \ge 0\) is considered. By suitably fixing the parameters \(\varrho\) and \(\varpi\) , in \(|a_n|\) and using these coefficients we discuss their application to enhance medical images in image processing. Furthermore, we introduce a novel image sharpening methodology that leverages coefficient bounds derived from the Miller-Ross function with fixing the parameters \(\varrho\) and \(\varpi\) values, integrating these bounds into pre-processing strategies for enhancing medical CT images.
