Artificial Intelligence Networks Combining Histopathology and Machine Learning Can Extract Axon Pathology in Autism Spectrum Disorder / Arash YAZDANBAKHSH in Autism Research, 18-11 (November 2025)  

Public ISBD [article]  inAutism Research > 18-11 (November 2025) . - p.2210-2230 Titre : Artificial Intelligence Networks Combining Histopathology and Machine Learning Can Extract Axon Pathology in Autism Spectrum Disorder Type de document : texte imprimé Auteurs : Arash YAZDANBAKHSH, Auteur ; Kim T.M. DANG, Auteur ; Kelvin KUANG, Auteur ; Tingru LIAN, Auteur ; Xuefeng LIU, Auteur ; Songlin XIE, Auteur ; Basilis ZIKOPOULOS, Auteur Article en page(s) : p.2210-2230 Langues : Anglais (eng) Mots-clés : anterior cingulate cortex  convolutional neural network  deep neural network  long-range pathways  short-range pathways  white matter Index. décimale : PER Périodiques Résumé : ABSTRACT Axon features that underlie the structural and functional organization of cortical pathways have distinct patterns in the brains of neurotypical controls (CTR) compared to individuals with Autism Spectrum Disorder (ASD). However, detailed axon study demands labor-intensive surveys and time-consuming analysis of microscopic sections from postmortem human brain tissue, making it challenging to systematically examine large regions of the brain. To address these challenges, we developed an approach that uses machine learning to automatically classify microscopic sections from ASD and CTR brains, while also considering different white matter regions: superficial white matter (SWM), which contains a majority of axons that connect nearby cortical areas, and deep white matter (DWM), which is comprised exclusively of axons that participate in long-range pathways. The result was a deep neural network that can successfully classify the white matter below the anterior cingulate cortex (ACC) of ASD and CTR groups with 98% accuracy, while also distinguishing between DWM and SWM pathway composition with high average accuracy, up to 80%. Examination of image regions important for network classification and misclassification, through sensitivity maps, along with multidimensional scaling analysis, helped identify key pathological markers of ASD and highlighted the spectrum of ASD heterogeneity and overlaps with neurotypical characteristics. Large datasets that can be used to expand training, validation, and testing of this network have the potential to automate high-resolution microscopic analysis of postmortem brain tissue, so that it can be used to systematically study white matter across brain regions in health and disease. En ligne : https://doi.org/10.1002/aur.70135 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=571 [article] Artificial Intelligence Networks Combining Histopathology and Machine Learning Can Extract Axon Pathology in Autism Spectrum Disorder [texte imprimé] / Arash YAZDANBAKHSH, Auteur ; Kim T.M. DANG, Auteur ; Kelvin KUANG, Auteur ; Tingru LIAN, Auteur ; Xuefeng LIU, Auteur ; Songlin XIE, Auteur ; Basilis ZIKOPOULOS, Auteur . - p.2210-2230. Langues : Anglais (eng) in Autism Research > 18-11 (November 2025) . - p.2210-2230 Mots-clés : anterior cingulate cortex  convolutional neural network  deep neural network  long-range pathways  short-range pathways  white matter Index. décimale : PER Périodiques Résumé : ABSTRACT Axon features that underlie the structural and functional organization of cortical pathways have distinct patterns in the brains of neurotypical controls (CTR) compared to individuals with Autism Spectrum Disorder (ASD). However, detailed axon study demands labor-intensive surveys and time-consuming analysis of microscopic sections from postmortem human brain tissue, making it challenging to systematically examine large regions of the brain. To address these challenges, we developed an approach that uses machine learning to automatically classify microscopic sections from ASD and CTR brains, while also considering different white matter regions: superficial white matter (SWM), which contains a majority of axons that connect nearby cortical areas, and deep white matter (DWM), which is comprised exclusively of axons that participate in long-range pathways. The result was a deep neural network that can successfully classify the white matter below the anterior cingulate cortex (ACC) of ASD and CTR groups with 98% accuracy, while also distinguishing between DWM and SWM pathway composition with high average accuracy, up to 80%. Examination of image regions important for network classification and misclassification, through sensitivity maps, along with multidimensional scaling analysis, helped identify key pathological markers of ASD and highlighted the spectrum of ASD heterogeneity and overlaps with neurotypical characteristics. Large datasets that can be used to expand training, validation, and testing of this network have the potential to automate high-resolution microscopic analysis of postmortem brain tissue, so that it can be used to systematically study white matter across brain regions in health and disease. En ligne : https://doi.org/10.1002/aur.70135 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=571

Imbalance of laminar-specific excitatory and inhibitory circuits of the orbitofrontal cortex in autism / Xuefeng LIU in Molecular Autism, 11 (2020)  

Public ISBD [article]  inMolecular Autism > 11 (2020) . - 83 p. Titre : Imbalance of laminar-specific excitatory and inhibitory circuits of the orbitofrontal cortex in autism Type de document : texte imprimé Auteurs : Xuefeng LIU, Auteur ; Julied BAUTISTA, Auteur ; Edward LIU, Auteur ; Basilis ZIKOPOULOS, Auteur Article en page(s) : 83 p. Langues : Anglais (eng) Mots-clés : Amygdala  Calbindin  Calretinin  Cortical layers  Emotions  Inhibitory neurons  Myelinated axons  Parvalbumin  Prefrontal cortex  Social interactions Index. décimale : PER Périodiques Résumé : BACKGROUND: The human orbitofrontal cortex (OFC) is involved in assessing the emotional significance of events and stimuli, emotion-based learning, allocation of attentional resources, and social cognition. Little is known about the structure, connectivity and excitatory/inhibitory circuit interactions underlying these diverse functions in human OFC, as well as how the circuit is disrupted in individuals with autism spectrum disorder (ASD). METHODS: We used post-mortem brain tissue from neurotypical adults and individuals with ASD. We examined the morphology and distribution of myelinated axons across cortical layers in OFC, at the single axon level, as a proxy of excitatory pathways. In the same regions, we also examined the laminar distribution of all neurons and neurochemically- and functionally-distinct inhibitory neurons that express the calcium-binding proteins parvalbumin (PV), calbindin (CB), and calretinin (CR). RESULTS: We found that the density of myelinated axons increased consistently towards layer 6, while the average axon diameter did not change significantly across layers in both groups. However, both the density and diameter of myelinated axons were significantly lower in the ASD group compared with the Control group. The distribution pattern and density of the three major types of inhibitory neurons was comparable between groups, but there was a significant reduction in the density of excitatory neurons across OFC layers in ASD. LIMITATIONS: This study is limited by the availability of human post-mortem tissue optimally processed for high-resolution microscopy and immunolabeling, especially from individuals with ASD. CONCLUSIONS: The balance between excitation and inhibition in OFC is at the core of its function, assessing and integrating emotional and social cues with internal states and external inputs. Our preliminary results provide evidence for laminar-specific changes in the ratio of excitation/inhibition in OFC of adults with ASD, with an overall weakening and likely disorganization of excitatory signals and a relative strengthening of local inhibition. These changes likely underlie pathology of major OFC communications with limbic or other cortices and the amygdala in individuals with ASD, and may provide the anatomic basis for disrupted transmission of signals for social interactions and emotions in autism. En ligne : http://dx.doi.org/10.1186/s13229-020-00390-x Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=433 [article] Imbalance of laminar-specific excitatory and inhibitory circuits of the orbitofrontal cortex in autism [texte imprimé] / Xuefeng LIU, Auteur ; Julied BAUTISTA, Auteur ; Edward LIU, Auteur ; Basilis ZIKOPOULOS, Auteur . - 83 p. Langues : Anglais (eng) in Molecular Autism > 11 (2020) . - 83 p. Mots-clés : Amygdala  Calbindin  Calretinin  Cortical layers  Emotions  Inhibitory neurons  Myelinated axons  Parvalbumin  Prefrontal cortex  Social interactions Index. décimale : PER Périodiques Résumé : BACKGROUND: The human orbitofrontal cortex (OFC) is involved in assessing the emotional significance of events and stimuli, emotion-based learning, allocation of attentional resources, and social cognition. Little is known about the structure, connectivity and excitatory/inhibitory circuit interactions underlying these diverse functions in human OFC, as well as how the circuit is disrupted in individuals with autism spectrum disorder (ASD). METHODS: We used post-mortem brain tissue from neurotypical adults and individuals with ASD. We examined the morphology and distribution of myelinated axons across cortical layers in OFC, at the single axon level, as a proxy of excitatory pathways. In the same regions, we also examined the laminar distribution of all neurons and neurochemically- and functionally-distinct inhibitory neurons that express the calcium-binding proteins parvalbumin (PV), calbindin (CB), and calretinin (CR). RESULTS: We found that the density of myelinated axons increased consistently towards layer 6, while the average axon diameter did not change significantly across layers in both groups. However, both the density and diameter of myelinated axons were significantly lower in the ASD group compared with the Control group. The distribution pattern and density of the three major types of inhibitory neurons was comparable between groups, but there was a significant reduction in the density of excitatory neurons across OFC layers in ASD. LIMITATIONS: This study is limited by the availability of human post-mortem tissue optimally processed for high-resolution microscopy and immunolabeling, especially from individuals with ASD. CONCLUSIONS: The balance between excitation and inhibition in OFC is at the core of its function, assessing and integrating emotional and social cues with internal states and external inputs. Our preliminary results provide evidence for laminar-specific changes in the ratio of excitation/inhibition in OFC of adults with ASD, with an overall weakening and likely disorganization of excitatory signals and a relative strengthening of local inhibition. These changes likely underlie pathology of major OFC communications with limbic or other cortices and the amygdala in individuals with ASD, and may provide the anatomic basis for disrupted transmission of signals for social interactions and emotions in autism. En ligne : http://dx.doi.org/10.1186/s13229-020-00390-x Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=433

---

1 (1 - 2 / 2) Par page : 25 50 100 200