Recent advances in machine learning and artificial intelligence technology have ensured automated evaluation of medical images. As a result, quantifiable diagnostic and prognostic biomarkers have been created. We discuss radiomics applications for the head and neck region in this paper. Molecular characterization, categorization, prognosis and therapy recommendation are given special consideration. In a narrative manner, we outline the fundamental technological principles, the overall idea and usual workflow of radiomic analysis and what seem to be the present and potential challenges in normal clinical practice. Clinical oncology intends for all of this to ensure informed decision support for personalized and useful cancer treatment. Head and neck cancers present a unique set of diagnostic and therapeutic challenges. These challenges are brought on by the complicated anatomy and heterogeneity of the area under investigation. Radiomics has the potential to address these barriers. Future research must be interdisciplinary and focus on the study of certain oncologic functions and outcomes, with external validation and multi-institutional cooperation in order to achieve this.

The sellar region represents a complex anatomical area, composed of multiple structures of different embryological derivation, including the skull base and the pituitary gland, along with vascular, nervous, and meningeal structures. Masses arising in this region include benign and malignant lesions arising from the pituitary gland itself, but also from vestigial embryological residues or surrounding tissues, that may require different therapeutic approaches. While assessing sellar region masses, the combination of clinical presentation and imaging features is fundamental to define hypotheses about their nature. MR represents the imaging modality of choice, providing information about the site of the lesion, its imaging features, and relation with adjacent structures, while CT is useful to confirm the presence of lesion calcifications or to reveal tumor invasion of bony structures. The aim of this pictorial review is to provide an overview of the common neoplasms and tumor-like conditions of the sellar region, according to the 2021 WHO Classification of Tumors of the Central Nervous System (fifth edition), with an emphasis on the radiologic-pathologic correlation. After a brief introduction on the anatomy of this region and the imaging and pathological techniques currently used, the most relevant MRI characteristics, clinical findings, and pathological data, including histologic and molecular features, will be shown and discussed, with the aim of facilitating an appropriate differential diagnosis among these entities.

Although multiple sclerosis (MS) can be conceptualized as a network disorder, brain network analyses typically require advanced MRI sequences not commonly acquired in clinical practice. Using conventional MRI, we assessed cross-sectional and longitudinal structural disconnection and morphometric similarity networks in people with MS (pwMS), along with their relationship with clinical disability. In this longitudinal monocentric study, 3T structural MRI of pwMS and healthy controls (HC) was retrospectively analyzed. Physical and cognitive disabilities were assessed with the expanded disability status scale (EDSS) and the symbol digit modalities test (SDMT), respectively. Demyelinating lesions were automatically segmented, and the corresponding masks were used to assess pairwise structural disconnection between atlas-defined brain regions based on normative tractography data. Using the Morphometric Inverse Divergence method, we computed morphometric similarity between cortical regions based on FreeSurfer surface reconstruction. Using network-based statistics (NBS) and its extension NBS-predict, we tested whether subject-level connectomes were associated with disease status, progression, clinical disability, and long-term confirmed disability progression (CDP), independently from global lesion burden and atrophy. We studied 461 pwMS (age = 37.2 ± 10.6 years, F/M = 324/137), corresponding to 1,235 visits (mean follow-up time = 1.9 ± 2.0 years, range = 0.1-13.3 years), and 55 HC (age = 42.4 ± 15.7 years; F/M = 25/30). Long-term clinical follow-up was available for 285 pwMS (mean follow-up time = 12.4 ± 2.8 years), 127 of whom (44.6%) exhibited CDP. At baseline, structural disconnection in pwMS was mostly centered around the thalami and cortical sensory and association hubs, while morphometric similarity was extensively disrupted (p_FWE < 0.01). EDSS was related to frontothalamic disconnection (p_FWE < 0.01) and disrupted morphometric similarity around the left perisylvian cortex (p_FWE = 0.02), while SDMT was associated with cortico-subcortical disconnection in the left hemisphere (p_FWE < 0.01). Longitudinally, both structural disconnection and morphometric similarity disruption significantly progressed (p_FWE = 0.04 and p_FWE < 0.01), correlating with EDSS increase (ρ = 0.07, p = 0.02 and ρ = 0.11, p < 0.001), while baseline disconnection predicted long-term CDP (accuracy = 59% [58-60], p = 0.03). Structural disconnection and morphometric similarity networks, as assessed through conventional MRI, are sensitive to MS-related brain damage and its progression. They explain disease-related clinical disability and predict its long-term evolution independently from global lesion burden and atrophy, potentially adding to established MRI measures as network-based biomarkers of disease severity and progression.

The complexity in diagnosing hereditary degenerative ataxias lies not only in their rarity, but also in the variety of different genetic conditions that can determine sometimes similar and overlapping clinical findings. In this light, Magnetic Resonance Imaging (MRI) plays a key role in the evaluation of these conditions, being a fundamental diagnostic tool needed not only to exclude other causes determining the observed clinical phenotype, but also to proper guide to an adequate genetic testing. Here, we propose an MRI-based diagnostic algorithm named CHARON (Characterization of Hereditary Ataxias Relying On Neuroimaging), to help in disentangling among the numerous, and apparently very similar, hereditary degenerative ataxias. Being conceived from a neuroradiological standpoint, it is based primarily on an accurate evaluation of the observed MRI findings, with the first and most important being the pattern of cerebellar atrophy. Along with the evaluation of the presence, or absence, of additional signal changes and/or supratentorial involvement, CHARON allows for the identification of a small groups of ataxias sharing similar imaging features. The integration of additional MRI findings, demographic, clinical and laboratory data allow then for the identification of typical, and in some cases pathognomonic, phenotypes of hereditary ataxias.

Background: Autosomal recessive spastic ataxia of Charlevoix‐Saguenay (ARSACS) and hereditary spastic paraplegia type 7 (SPG7) represent the most common genotypes of spastic ataxia (SPAX). To date, their magnetic resonance imaging (MRI) features have only been described qualitatively, and a pure neuroradiological differential diagnosis between these two conditions is difficult to achieve.Objectives: To test the performance of MRI measures to discriminate between ARSACS and SPG7 (as an index of common SPAX disease).Methods: In this prospective multicenter study, 3D‐T1‐weighted images of 59 ARSACS (35.4 ± 10.3 years, M/F = 33/26) and 78 SPG7 (54.8 ± 10.3 years, M/F = 51/27) patients of the PROSPAX Consortium were analyzed, together with 30 controls (45.9 ± 16.9 years, M/F = 15/15). Different linear and surface measures were evaluated. A receiver operating characteristic analysis was performed, calculating area under the curve (AUC) and corresponding diagnostic accuracy parameters.Results: The pons area proved to be the only metric increased exclusively in ARSACS patients (P = 0.02). Other different measures were reduced in ARSACS and SPG7 compared with controls (all with P ≤ 0.005). A cut‐off value equal to 1.67 of the pons‐to‐superior vermis area ratio proved to have the highest AUC (0.98, diagnostic accuracy 93%, sensitivity 97%) in discriminating between ARSACS and SPG7.Conclusions: Evaluation of the pons‐to‐superior vermis area ratio can discriminate ARSACS from other SPAX patients, as exemplified here by SPG7. Hence, we hereby propose this ratio as the Magnetic Resonance Index for the Assessment and Recognition of patients harboring SACS mutations (MRI‐ARSACS), a novel diagnostic tool able to identify ARSACS patients and useful for discriminating ARSACS from other SPAX patients undergoing MRI. © 2024 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Background: Autosomal Recessive Spastic Ataxia of Charlevoix-Saguenay (ARSACS) and Spastic Paraplegia Type 7 (SPG7) are paradigmatic spastic ataxias (SPAX) with suggested white matter (WM) involvement. Aim of this work was to thoroughly disentangle the degree of WM involvement in these conditions, evaluating both macrostructure and microstructure via the analysis of diffusion MRI (dMRI) data. Material and methods: In this multi-center prospective study, ARSACS and SPG7 patients and Healthy Controls (HC) were enrolled, all undergoing a standardized dMRI protocol and a clinimetrics evaluation including the Scale for the Assessment and Rating of Ataxia (SARA). Differences in terms of WM volume or global microstructural WM metrics were probed, as well as the possible occurrence of a spatially defined microstructural WM involvement via voxel-wise analyses, and its correlation with patients’ clinical status. Results: Data of 37 ARSACS (M/F = 21/16; 33.4 ± 12.4 years), 37 SPG7 (M/F = 24/13; 55.7 ± 10.7 years), and 29 HC (M/F = 13/16; 42.1 ± 17.2 years) were analyzed. While in SPG7, only a mild mean microstructural damage was found compared to HC, ARSACS patients present a severe WM involvement, with a reduced global volume (p < 0.001), an alteration of all microstructural metrics (all with p < 0.001), without a spatially defined pattern of damage but with a prominent involvement of commissural fibers. Finally, in ARSACS, a correlation between microstructural damage and SARA scores was found (p = 0.004). Conclusion: In ARSACS, but not SPG7 patients, we observed a complex and multi-faced involvement of brain WM, with a clinically meaningful widespread loss of axonal and dendritic integrity, secondary demyelination and, overall, a reduction in cellularity and volume.