A multi-modal parcellation of the human cerebral cortex was developed using data from the Human Connectome Project (HCP) and an objective semi-automated neuroanatomical approach. The study identified 180 cortical areas per hemisphere, delineated by sharp changes in cortical architecture, function, connectivity, and topography. The parcellation included 97 new areas and 83 previously reported areas. A machine-learning classifier was trained to recognize the multi-modal 'fingerprint' of each cortical area, enabling automated delineation in new subjects. The parcellation and classifier are freely available, offering improved neuroanatomical precision for studies of the structural and functional organization of the human cerebral cortex. The study combined four properties—architecture, function, connectivity, and topography—to define cortical areas. Architectural measures were derived from T1w and T2w structural images, while functional data came from task and resting-state fMRI. The parcellation was validated using two independent groups of HCP subjects, showing high reproducibility across modalities. The parcellation was found to be highly consistent across subjects, with strong correlations between group averages and individual data. The parcellation included 180 areas, with some areas being complexes of multiple subareas. The parcellation was validated using cross-validation techniques, showing that 96.6% of areas were correctly identified in new subjects. The parcellation was also tested for its ability to detect atypical areal patterns in some individuals, demonstrating its robustness in identifying both typical and atypical areas. The study also developed an automated method for individual subject parcellation using a machine learning classifier. This classifier was trained on the 210P group and tested on the 210V group, showing high accuracy in detecting cortical areas. The classifier was able to replicate the group parcellation and correctly identify atypical areas in some subjects. The study highlights the importance of multi-modal data in defining cortical areas and provides a framework for future studies. The parcellation and classifier are freely available, enabling further research into the structural and functional organization of the human cerebral cortex. The study also emphasizes the need for improved neuroanatomical methods to better understand the brain's organization and function.A multi-modal parcellation of the human cerebral cortex was developed using data from the Human Connectome Project (HCP) and an objective semi-automated neuroanatomical approach. The study identified 180 cortical areas per hemisphere, delineated by sharp changes in cortical architecture, function, connectivity, and topography. The parcellation included 97 new areas and 83 previously reported areas. A machine-learning classifier was trained to recognize the multi-modal 'fingerprint' of each cortical area, enabling automated delineation in new subjects. The parcellation and classifier are freely available, offering improved neuroanatomical precision for studies of the structural and functional organization of the human cerebral cortex. The study combined four properties—architecture, function, connectivity, and topography—to define cortical areas. Architectural measures were derived from T1w and T2w structural images, while functional data came from task and resting-state fMRI. The parcellation was validated using two independent groups of HCP subjects, showing high reproducibility across modalities. The parcellation was found to be highly consistent across subjects, with strong correlations between group averages and individual data. The parcellation included 180 areas, with some areas being complexes of multiple subareas. The parcellation was validated using cross-validation techniques, showing that 96.6% of areas were correctly identified in new subjects. The parcellation was also tested for its ability to detect atypical areal patterns in some individuals, demonstrating its robustness in identifying both typical and atypical areas. The study also developed an automated method for individual subject parcellation using a machine learning classifier. This classifier was trained on the 210P group and tested on the 210V group, showing high accuracy in detecting cortical areas. The classifier was able to replicate the group parcellation and correctly identify atypical areas in some subjects. The study highlights the importance of multi-modal data in defining cortical areas and provides a framework for future studies. The parcellation and classifier are freely available, enabling further research into the structural and functional organization of the human cerebral cortex. The study also emphasizes the need for improved neuroanatomical methods to better understand the brain's organization and function.