CapNet: An Automatic Attention-Based with Mixer Model for Cardiovascular Magnetic Resonance Image Segmentation.

Deep neural networks have shown excellent performance in medical image segmentation, especially for cardiac images. Transformer-based models, though having advantages over convolutional neural networks due to the ability of long-range dependence learning, still have shortcomings such as having a large number of parameters and and high computational cost. Additionally, for better results, they are often pretrained on a larger data, thus requiring large memory size and increasing resource expenses. In this study, we propose a new lightweight but efficient model, namely CapNet, based on convolutions and mixing modules for cardiac segmentation from magnetic resonance images (MRI) that can be trained from scratch with a small amount of parameters. To handle varying sizes and shapes which often occur in cardiac systolic and diastolic phases, we propose attention modules for pooling, spatial, and channel information. We also propose a novel loss called the Tversky Shape Power Distance function based on the shape dissimilarity between labels and predictions that shows promising performances compared to other losses. Experiments on three public datasets including ACDC benchmark, Sunnybrook data, and MS-CMR challenge are conducted and compared with other state of the arts (SOTA). For binary segmentation, the proposed CapNet obtained the Dice similarity coefficient (DSC) of 94% and 95.93% for respectively the Endocardium and Epicardium regions with Sunnybrook dataset, 94.49% for Endocardium, and 96.82% for Epicardium with the ACDC data. Regarding the multiclass case, the average DSC by CapNet is 93.05% for the ACDC data; and the DSC scores for the MS-CMR are 94.59%, 92.22%, and 93.99% for respectively the bSSFP, T2-SPAIR, and LGE sequences of the MS-CMR. Moreover, the statistical significance analysis tests with p-value < 0.05 compared with transformer-based methods and some CNN-based approaches demonstrated that the CapNet, though having fewer training parameters, is statistically significant. The promising evaluation metrics show comparative results in both Dice and IoU indices compared to SOTA CNN-based and Transformer-based architectures.

Altered Heschl's gyrus duplication pattern in first-episode schizophrenia.

BACKGROUND: Reduced gray matter volumes in the superior temporal gyrus and its subregions, such as Heschl's gyrus (HG) and the planum temporale (PT), have been reported in schizophrenia (Sz). However, it remains unclear whether patients exhibit an altered sulcogyral pattern on the superior temporal plane. METHODS: This magnetic resonance imaging study examined the distribution of HG duplication patterns [i.e., single HG, common stem duplication (CSD), or complete posterior duplication (CPD)] and their relationships with clinical variables and gray matter volumes in the HG and PT of 64 first-episode (FE) patients with Sz and 64 healthy controls. RESULTS: The prevalence of duplicated HG patterns was significantly higher and gray matter volumes in the HG and PT of both hemispheres were smaller in FESz patients than in healthy controls. The right CPD pattern in the FESz group was associated with less severe positive symptoms. In the FESz and control groups, CSD and CPD patterns correlated with larger volumes in the HG and PT, respectively. CONCLUSION: The present results revealed an altered HG duplication pattern at the earliest phase of Sz, which may reflect early neurodevelopmental anomalies. However, reduced HG and PT volumes in the FESz were not explained by this sulcogyral pattern only, supporting the complex superior temporal pathology of Sz.

Potential contribution of pineal atrophy and pineal cysts toward vulnerability and clinical characteristics of psychosis.

BACKGROUND: Magnetic resonance imaging (MRI) studies reported pineal gland atrophy in schizophrenia patients and individuals at a clinical high risk of developing psychosis, implicating abnormalities in melatonin secretion in the pathophysiology of psychosis. However, it currently remains unclear whether the morphology of the pineal gland contributes to symptomatology and sociocognitive functions. METHODS: This MRI study examined pineal gland volumes and the prevalence of pineal cysts as well as their relationship with clinical characteristics in 57 at risk mental state (ARMS) subjects, 63 patients with schizophrenia, and 61 healthy controls. The Social and Occupational Functioning Assessment Scale (SOFAS), the Schizophrenia Cognition Rating Scale (SCoRS), and the Brief Assessment of Cognition in Schizophrenia (BACS) were used to assess sociocognitive functions, while the Positive and Negative Syndrome Scale was employed to evaluate clinical symptoms in ARMS subjects and schizophrenia patients. RESULTS: Pineal gland volumes were significantly smaller in the ARMS and schizophrenia groups than in the controls, while no significant differences were observed in the prevalence of pineal cysts. Although BACS, SCoRS, and SOFAS scores were not associated with pineal morphology, patients with pineal cysts in the schizophrenia group exhibited severe positive psychotic symptoms with rather mild negative symptoms. CONCLUSION: The present results indicate the potential of pineal atrophy as a vulnerability marker in various stages of psychosis and suggest that pineal cysts influence the clinical subtype of schizophrenia.

Heschl's Gyrus Duplication Pattern in Individuals at Risk of Developing Psychosis and Patients With Schizophrenia.

An increased prevalence of duplicated Heschl's gyrus (HG), which may reflect an early neurodevelopmental pathology, has been reported in schizophrenia (Sz). However, it currently remains unclear whether individuals at risk of psychosis exhibit similar brain morphological characteristics. This magnetic resonance imaging study investigated the distribution of HG gyrification patterns [i.e., single HG, common stem duplication (CSD), and complete posterior duplication (CPD)] and their relationship with clinical characteristics in 57 individuals with an at-risk mental state (ARMS) [of whom 5 (8.8%) later developed Sz], 63 patients with Sz, and 61 healthy comparisons. The prevalence of duplicated HG patterns (i.e., CSD or CPD) bilaterally was significantly higher in the ARMS and Sz groups than in the controls, whereas no significant differences were observed in HG patterns between these groups. The left CSD pattern, particularly in the Sz group, was associated with a verbal fluency deficit. In the ARMS group, left CSD pattern was related to a more severe general psychopathology. The present results suggest that an altered gyrification pattern on the superior temporal plane reflects vulnerability factors associated with Sz, which may also contribute to the clinical features of high-risk individuals, even without the onset of psychosis.

Increased Heschl's Gyrus Duplication in Schizophrenia Spectrum Disorders: A Cross-Sectional MRI Study.

Duplicated Heschl's gyrus (HG) is prevalent in patients with schizophrenia and may reflect early neurodevelopmental anomalies. However, it currently remains unclear whether patients with schizotypal disorder, a prototypic disorder within the schizophrenia spectrum, exhibit a similar HG gyrification pattern. In this magnetic resonance imaging study, HG gyrification patterns were examined in 47 patients with schizotypal disorder, 111 with schizophrenia, and 88 age- and sex-matched healthy subjects. HG gyrification patterns were classified as single, common stem duplication (CSD), or complete posterior duplication (CPD). The prevalence of the duplicated HG patterns (CSD or CPD) bilaterally was higher in the schizophrenia and schizotypal groups than in healthy controls, whereas no significant difference was observed between the schizophrenia and schizotypal groups. Schizophrenia patients with the right CPD pattern had less severe positive symptoms, whereas the right single HG pattern was associated with higher doses of antipsychotic medication in schizotypal patients. The present study demonstrated shared HG gyrification patterns in schizophrenia spectrum disorders, which may reflect a common biological vulnerability factor. HG patterns may also be associated with susceptibility to psychopathology.

Longitudinal Changes in Brain Gyrification in Schizophrenia Spectrum Disorders.

Previous magnetic resonance imaging (MRI) studies reported increased brain gyrification in schizophrenia and schizotypal disorder, a prototypic disorder within the schizophrenia spectrum. This may reflect deviations in early neurodevelopment; however, it currently remains unclear whether the gyrification pattern longitudinally changes over the course of the schizophrenia spectrum. The present MRI study using FreeSurfer compared longitudinal changes (mean inter-scan interval of 2.7 years) in the local gyrification index (LGI) in the entire cortex among 23 patients with first-episode schizophrenia, 14 with schizotypal disorder, and 39 healthy controls. Significant differences were observed in longitudinal LGI changes between these groups; the schizophrenia group exhibited a progressive decline in LGI, predominantly in the fronto-temporal regions, whereas LGI increased over time in several brain regions in the schizotypal and control groups. In the schizophrenia group, a greater reduction in LGI over time in the right precentral and post central regions correlated with smaller improvements in negative symptoms during the follow-up period. The cumulative medication dosage during follow-up negatively correlated with a longitudinal LGI increase in the right superior parietal area in the schizotypal group, but did not affect longitudinal LGI changes in the schizophrenia group. Collectively, these results suggest that gyrification patterns in the schizophrenia spectrum reflect both early neurodevelopmental abnormalities as a vulnerability factor and active brain pathology in the early stages of schizophrenia.