Medical Image Processing is pivotal for enhancing precision in clinical diagnostics, particularly in Nuclear Medicine (NM). Thyroid scintigraphy with 99mTc plays a crucial role in thyroid function assessment. However, these images often suffer from artifacts and noise, such as Poisson noise, hindering accurate diagnosis. This study focuses on assessing Poisson noise reduction in thyroid scintigraphy using three specific filters: the OCCO morphological filter, Wavelet filter, and Wiener filter. The study involved nine thyroid scintigraphy images with Poisson noise. Evaluation included selecting optimal structuring elements for the OCCO filter, comparing various wavelet families, and assessing the Wiener filter's adaptive noise removal. Results indicated that, among structuring elements, the Diamond and Disk of size 1 yielded the best performance. The bior3.5 Wavelet family demonstrated superior results, while the Wiener filter, configured with adaptive noise parameters, outperformed other methods. In conclusion, this study offers insights into effective noise reduction techniques for thyroid scintigraphy, contributing to the advancement of image processing methodologies in medical diagnostics.