Spheroids are three-dimensional models that play a crucial role in the study of tissues and tumors. Advances in technology have enabled the automated generation of spheroids with various experimental parameters, but the manual analysis of such data is time-consuming and prone to inaccuracies. Therefore, a robust and rapid solution for the morphological analysis of these models is required. This study presents a Python-based algorithm for the quantified analysis of 3D tumor spheroids (PANC1 cell line) produced through a robotic-enabled platform. The pipeline includes sharp image detection, instance segmentation, and contour analysis, using a YOLO (You Only Look Once) machine learning model to identify key morphological features of the tumor models, such as their shape, area, and circularity. The model is custom-trained on a dataset comprising 518 images of 3D tumor spheroids. Its accuracy is validated by comparing its results with manual annotations performed by experts on the test dataset. The model achieved an F1 score of 0.872 in training results, indicating a strong balance between precision and recall in its classification of morphological features. Furthermore, the algorithm facilitates the rapid and reproducible analysis of large datasets, reducing the workload and improving the overall quality of morphological assessment. This contributes to better insights into tumor behavior and the effects of drug treatments.