RNA sequencing of peripheral blood in amyotrophic lateral sclerosis reveals distinct molecular subtypes: Considerations for biomarker discovery.

AIM: Amyotrophic lateral sclerosis (ALS) is a heterogeneous neurodegenerative disease with limited therapeutic options. A key factor limiting the development of effective therapeutics is the lack of disease biomarkers. We sought to assess whether biomarkers for diagnosis, prognosis or cohort stratification could be identified by RNA sequencing (RNA-seq) of ALS patient peripheral blood. METHODS: Whole blood RNA-seq data were generated for 96 Australian sporadic ALS (sALS) cases and 48 healthy controls (NCBI GEO accession GSE234297). Differences in sALS-control gene expression, transcript usage and predicted leukocyte proportions were assessed, with pathway analysis used to predict the activity state of biological processes. Weighted Gene Co-expression Network Analysis (WGCNA) and machine learning algorithms were applied to search for diagnostic and prognostic gene expression patterns. Unsupervised clustering analysis was employed to determine whether sALS patient subgroups could be detected. RESULTS: Two hundred and forty-five differentially expressed genes were identified in sALS patients relative to controls, with enrichment of immune, metabolic and stress-related pathways. sALS patients also demonstrated switches in transcript usage across a small set of genes. We established a classification model that distinguished sALS patients from controls with an accuracy of 78% (sensitivity: 79%, specificity: 75%) using the expression of 20 genes. Clustering analysis identified four patient subgroups with gene expression signatures and immune cell proportions reflective of distinct peripheral effects. CONCLUSIONS: Our findings suggest that peripheral blood RNA-seq can identify diagnostic biomarkers and distinguish molecular subtypes of sALS patients however, its prognostic value requires further investigation.

Inter-Regional Proteomic Profiling of the Human Brain Using an Optimized Protein Extraction Method from Formalin-Fixed Tissue to Identify Signaling Pathways.

Proteomics offers vast potential for studying the molecular regulation of the human brain. Formalin fixation is a common method for preserving human tissue; however, it presents challenges for proteomic analysis. In this study, we compared the efficiency of two different protein-extraction buffers on three post-mortem, formalin-fixed human brains. Equal amounts of extracted proteins were subjected to in-gel tryptic digestion and LC-MS/MS. Protein, peptide sequence, and peptide group identifications; protein abundance; and gene ontology pathways were analyzed. Protein extraction was superior using lysis buffer containing tris(hydroxymethyl)aminomethane hydrochloride, sodium dodecyl sulfate, sodium deoxycholate, and Triton X-100 (TrisHCl, SDS, SDC, Triton X-100), which was then used for inter-regional analysis. Pre-frontal, motor, temporal, and occipital cortex tissues were analyzed by label free quantification (LFQ) proteomics, Ingenuity Pathway Analysis and PANTHERdb. Inter-regional analysis revealed differential enrichment of proteins. We found similarly activated cellular signaling pathways in different brain regions, suggesting commonalities in the molecular regulation of neuroanatomically-linked brain functions. Overall, we developed an optimized, robust, and efficient method for protein extraction from formalin-fixed human brain tissue for in-depth LFQ proteomics. We also demonstrate herein that this method is suitable for rapid and routine analysis to uncover molecular signaling pathways in the human brain.

Use of deep learning in the MRI diagnosis of Chiari malformation type I.

PURPOSE: To train deep learning convolutional neural network (CNN) models for classification of clinically significant Chiari malformation type I (CM1) on MRI to assist clinicians in diagnosis and decision making. METHODS: A retrospective MRI dataset of patients diagnosed with CM1 and healthy individuals with normal brain MRIs from the period January 2010 to May 2020 was used to train ResNet50 and VGG19 CNN models to automatically classify images as CM1 or normal. A total of 101 patients diagnosed with CM1 requiring surgery and 111 patients with normal brain MRIs were included (median age 30 with an interquartile range of 23-43; 81 women with CM1). Isotropic volume transformation, image cropping, skull stripping, and data augmentation were employed to optimize model accuracy. K-fold cross validation was used to calculate sensitivity, specificity, and the area under receiver operating characteristic curve (AUC) for model evaluation. RESULTS: The VGG19 model with data augmentation achieved a sensitivity of 97.1% and a specificity of 97.4% with an AUC of 0.99. The ResNet50 model achieved a sensitivity of 94.0% and a specificity of 94.4% with an AUC of 0.98. CONCLUSIONS: VGG19 and ResNet50 CNN models can be trained to automatically detect clinically significant CM1 on MRI with a high sensitivity and specificity. These models have the potential to be developed into clinical support tools in diagnosing CM1.

Foundations of Multiparametric Brain Tumour Imaging Characterisation Using Machine Learning.

The heterogeneity of brain tumours at the molecular, metabolic and structural levels poses significant challenge for accurate tissue characterisation. Artificial intelligence and radiomics have emerged as valuable tools to analyse quantitative features extracted from medical images which capture the complex microenvironment of brain tumours. In particular, a number of computational tools including machine learning algorithms have been proposed for image preprocessing, tumour segmentation, feature extraction, classification, and prognostic stratifications as well. In this chapter, we explore the fundamentals of multiparametric brain tumour characterisation, as an understanding of the strengths, limitations and applications of these tools allows clinicians to better develop and evaluate models with improved diagnostic and prognostic value in brain tumour patients.

Medical Image Classification Based on Semi-Supervised Generative Adversarial Network and Pseudo-Labelling.

Deep learning has substantially improved the state-of-the-art in object detection and image classification. Deep learning usually requires large-scale labelled datasets to train the models; however, due to the restrictions in medical data sharing and accessibility and the expensive labelling cost, the application of deep learning in medical image classification has been dramatically hindered. In this study, we propose a novel method that leverages semi-supervised adversarial learning and pseudo-labelling to incorporate the unlabelled images in model learning. We validate the proposed method on two public databases, including ChestX-ray14 for lung disease classification and BreakHis for breast cancer histopathological image diagnosis. The results show that our method achieved highly effective performance with an accuracy of 93.15% while using only 30% of the labelled samples, which is comparable to the state-of-the-art accuracy for chest X-ray classification; it also outperformed the current methods in multi-class breast cancer histopathological image classification with a high accuracy of 96.87%.

SoftMatch: Comparing Scanpaths Using Combinatorial Spatio-Temporal Sequences with Fractal Curves.

Recent studies matching eye gaze patterns with those of others contain research that is heavily reliant on string editing methods borrowed from early work in bioinformatics. Previous studies have shown string editing methods to be susceptible to false negative results when matching mutated genes or unordered regions of interest in scanpaths. Even as new methods have emerged for matching amino acids using novel combinatorial techniques, scanpath matching is still limited by a traditional collinear approach. This approach reduces the ability to discriminate between free viewing scanpaths of two people looking at the same stimulus due to the heavy weight placed on linearity. To overcome this limitation, we here introduce a new method called SoftMatch to compare pairs of scanpaths. SoftMatch diverges from traditional scanpath matching in two different ways: firstly, by preserving locality using fractal curves to reduce dimensionality from 2D Cartesian (x,y) coordinates into 1D (h) Hilbert distances, and secondly by taking a combinatorial approach to fixation matching using discrete Fréchet distance measurements between segments of scanpath fixation sequences. These matching "sequences of fixations over time" are a loose acronym for SoftMatch. Results indicate high degrees of statistical and substantive significance when scoring matches between scanpaths made during free-form viewing of unfamiliar stimuli. Applications of this method can be used to better understand bottom up perceptual processes extending to scanpath outlier detection, expertise analysis, pathological screening, and salience prediction.

Prediction of anxiety disorders using a feature ensemble based bayesian neural network.

Anxiety disorders are common among youth, posing risks to physical and mental health development. Early screening can help identify such disorders and pave the way for preventative treatment. To this end, the Youth Online Diagnostic Assessment (YODA) tool was developed and deployed to predict youth disorders using online screening questionnaires filled by parents. YODA facilitated collection of several novel unique datasets of self-reported anxiety disorder symptoms. Since the data is self-reported and often noisy, feature selection needs to be performed on the raw data to improve accuracy. However, a single set of selected features may not be informative enough. Consequently, in this work we propose and evaluate a novel feature ensemble based Bayesian Neural Network (FE-BNN) that exploits an ensemble of features for improving the accuracy of disorder predictions. We evaluate the performance of FE-BNN on three disorder-specific datasets collected by YODA. Our method achieved the AUC of 0.8683, 0.8769, 0.9091 for the predictions of Separation Anxiety Disorder, Generalized Anxiety Disorder and Social Anxiety Disorder, respectively. These results provide initial evidence that our method outperforms the original diagnostic scoring function of YODA and several other baseline methods for three anxiety disorders, which can practically help prioritizing diagnostic interviews. Our promising results call for investigation of interpretable methods maintaining high predictive accuracy.

Application of deep learning for automatic segmentation of brain tumors on magnetic resonance imaging: a heuristic approach in the clinical scenario.

PURPOSE: Accurate brain tumor segmentation on magnetic resonance imaging (MRI) has wide-ranging applications such as radiosurgery planning. Advances in artificial intelligence, especially deep learning (DL), allow development of automatic segmentation that overcome the labor-intensive and operator-dependent manual segmentation. We aimed to evaluate the accuracy of the top-performing DL model from the 2018 Brain Tumor Segmentation (BraTS) challenge, the impact of missing MRI sequences, and whether a model trained on gliomas can accurately segment other brain tumor types. METHODS: We trained the model using Medical Decathlon dataset, applied it to the BraTS 2019 glioma dataset, and developed additional models using individual and multimodal MRI sequences. The Dice score was calculated to assess the model's accuracy compared to ground truth labels by neuroradiologists on BraTS dataset. The model was then applied to a local dataset of 105 brain tumors, performance of which was qualitatively evaluated. RESULTS: The DL model using pre- and post-gadolinium contrast T1 and T2 FLAIR sequences performed best, with a Dice score 0.878 for whole tumor, 0.732 tumor core, and 0.699 active tumor. Lack of T1 or T2 sequences did not significantly degrade performance, but FLAIR and T1C were important contributors. All segmentations performed by the model in the local dataset, including non-glioma cases, were considered accurate by a pool of specialists. CONCLUSION: The DL model could use available MRI sequences to optimize glioma segmentation and adopt transfer learning to segment non-glioma tumors, thereby serving as a useful tool to improve treatment planning and personalized surveillance of patients.

Demyelination precedes axonal loss in the transneuronal spread of human neurodegenerative disease.

The spread of neurodegeneration through the human brain network is reported as underlying the progression of neurodegenerative disorders. However, the exact mechanisms remain unknown. The human visual pathway is characterized by its unique hierarchical architecture and, therefore, represents an ideal model to study trans-synaptic degeneration, in contrast to the complexity in neural connectivity of the whole brain. Here we show in two specifically selected patient cohorts, including (i) glaucoma patients with symmetrical bilateral hemifield defects respecting the horizontal meridian (n = 25, 14 females, 64.8 ± 10.1 years; versus 13 normal controls with similar age/sex distributions); and (ii) multiple sclerosis patients without optic radiation lesions (to avoid potential effects of lesions on diffusivity measures) (n = 30, 25 females, 37.9 ± 10.8 years; versus 20 controls), that there are measurable topographic changes in the posterior visual pathways corresponding to the primary optic nerve defects. A significant anisotropic increase of water diffusion was detected in both patient cohorts in the optic radiations, characterized by changes in perpendicular (radial) diffusivity (a measure of myelin integrity) that extended more posteriorly than those observed in parallel (axial) diffusivity (reflecting axonal integrity). In glaucoma, which is not considered a demyelinating disease, the observed increase in radial diffusivity within the optic radiations was validated by topographically linked delay of visual evoked potential latency, a functional measure of demyelination. Radial diffusivity change in the optic radiations was also associated with an asymmetrical reduction in the thickness of the calcarine cortex in glaucoma. In addition, 3 years longitudinal observation of the multiple sclerosis patient cohort revealed an anterograde increase of radial diffusivity in the anterior part of optic radiations which again was retinotopically associated with the primary damage caused by optic neuritis. Finally, in an animal model of optic nerve injury, we observed early glial activation and demyelination in the posterior visual projections, evidenced by the presence of myelin-laden macrophages. This occurred prior to the appearance of amyloid precursor protein accumulation, an indicator of disrupted fast axonal transport. This study demonstrated strong topographical spread of neurodegeneration along recognized neural projections and showed that myelin and glial pathology precedes axonal loss in the process, suggesting that the mechanism of trans-synaptic damage may be at least partially mediated by glial components at the cellular level. The findings may have broad biological and therapeutic implications for other neurodegenerative disorders.

Differing Structural and Functional Patterns of Optic Nerve Damage in Multiple Sclerosis and Neuromyelitis Optica Spectrum Disorder.

PURPOSE: To assess differential patterns of axonal loss and demyelination in the optic nerve in multiple sclerosis (MS) and neuromyelitis optica spectrum disorders (NMOSD). DESIGN: Cross-sectional study. PARTICIPANTS: One hundred ninety-two participants, including 136 MS patients (272 eyes), 19 NMOSD patients (38 eyes), and 37 healthy control participants (74 eyes). METHODS: All participants underwent spectral-domain OCT scans and multifocal visual evoked potential (mfVEP) recordings. High-resolution magnetic resonance imaging (MRI) with the diffusion protocol also was performed in all patients. MAIN OUTCOME MEASURES: Ganglion cell-inner plexiform layer (GCIPL) thickness and mfVEP amplitude and latency at 5 eccentricities; global and temporal retinal nerve fiber layer thickness. RESULTS: In optic neuritis (ON) eyes, the NMOSD patients had more severe GCIPL loss (P < 0.001) and mfVEP amplitude reduction (P < 0.001) compared with MS patients, whereas in contrast, mfVEP latency delay was more evident in MS patients (P < 0.001). The NMOSD patients showed more morphologic and functional loss at the foveal to parafoveal region, whereas the MS patients showed evenly distributed damage at the macula. Correlation analysis demonstrated a strong structure-function (OCT-mfVEP) association in the NMOSD patients, which was only moderate in the MS patients. In non-ON (NON) eyes, the MS patients showed significantly thinner GCIPL than controls (P < 0.001), whereas no GCIPL loss was observed in NON eyes in NMOSD. In addition, a significant correlation was found between all OCT and mfVEP measures in MS patients, but not in NMOSD patients. MRI demonstrated significant lesional load in the optic radiation in MS compared to NMOSD eyes (P = 0.002), which was related to the above OCT and mfVEP changes in NON eyes. CONCLUSIONS: Our study demonstrated different patterns of ON damage in NMOSD and MS. In MS, the ON damage was less severe, with demyelination as the main pathologic component, whereas in NMOSD, axonal loss was more severe compared with myelin loss. The disproportional mfVEP amplitude and latency changes suggested predominant axonal damage within the anterior visual pathway as the main clinical feature of NMOSD, in contrast to MS, where demyelination spreads along the entire visual pathway.

Structure of the mitochondrial ATP synthase from Pichia angusta determined by electron cryo-microscopy.

The structure of the intact monomeric ATP synthase from the fungus, Pichia angusta, has been solved by electron cryo-microscopy. The structure provides insights into the mechanical coupling of the transmembrane proton motive force across mitochondrial membranes in the synthesis of ATP. This mechanism requires a strong and integral stator, consisting of the catalytic α3ß3-domain, peripheral stalk, and, in the membrane domain, subunit a and associated supernumerary subunits, kept in contact with the rotor turning at speeds up to 350 Hz. The stator's integrity is ensured by robust attachment of both the oligomycin sensitivity conferral protein (OSCP) to the catalytic domain and the membrane domain of subunit b to subunit a. The ATP8 subunit provides an additional brace between the peripheral stalk and subunit a. At the junction between the OSCP and the apparently stiff, elongated α-helical b-subunit and associated d- and h-subunits, an elbow or joint allows the stator to bend to accommodate lateral movements during the activity of the catalytic domain. The stator may also apply lateral force to help keep the static a-subunit and rotating c10-ring together. The interface between the c10-ring and the a-subunit contains the transmembrane pathway for protons, and their passage across the membrane generates the turning of the rotor. The pathway has two half-channels containing conserved polar residues provided by a bundle of four α-helices inclined at ∼30° to the plane of the membrane, similar to those described in other species. The structure provides more insights into the workings of this amazing machine.

The purification and characterization of ATP synthase complexes from the mitochondria of four fungal species.

The ATP synthases have been isolated by affinity chromatography from the mitochondria of the fungal species Yarrowia lipolytica, Pichia pastoris, Pichia angusta and Saccharomyces cerevisiae. The subunit compositions of the purified enzyme complexes depended on the detergent used to solubilize and purify the complex, and the presence or absence of exogenous phospholipids. All four enzymes purified in the presence of n-dodecyl-ß-D-maltoside had a complete complement of core subunits involved directly in the synthesis of ATP, but they were deficient to different extents in their supernumerary membrane subunits. In contrast, the enzymes from P. angusta and S. cerevisiae purified in the presence of n-decyl-ß-maltose neopentyl glycol and the phospholipids 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine, cardiolipin (diphosphatidylglycerol) and 1-palmitoyl-2-oleoyl-sn-glycero-3-[phospho-rac-(1-glycerol)] had a complete complement of core subunits and also contained all of the known supernumerary membrane subunits, e, f, g, j, k and ATP8 (or Aap1), plus an additional new membrane component named subunit l, related in sequence to subunit k. The catalytic domain of the enzyme from P. angusta was more resistant to thermal denaturation than the enzyme from S. cerevisiae, but less stable than the catalytic domain of the bovine enzyme, but the stator and the integrity of the transmembrane proton pathway were most stable in the enzyme from P. angusta. The P. angusta enzyme provides a suitable source of enzyme for studying the structure of the membrane domain and properties associated with that sector of the enzyme complex.

Dictionary Pruning with Visual Word Significance for Medical Image Retrieval.

Content-based medical image retrieval (CBMIR) is an active research area for disease diagnosis and treatment but it can be problematic given the small visual variations between anatomical structures. We propose a retrieval method based on a bag-of-visual-words (BoVW) to identify discriminative characteristics between different medical images with Pruned Dictionary based on Latent Semantic Topic description. We refer to this as the PD-LST retrieval. Our method has two main components. First, we calculate a topic-word significance value for each visual word given a certain latent topic to evaluate how the word is connected to this latent topic. The latent topics are learnt, based on the relationship between the images and words, and are employed to bridge the gap between low-level visual features and high-level semantics. These latent topics describe the images and words semantically and can thus facilitate more meaningful comparisons between the words. Second, we compute an overall-word significance value to evaluate the significance of a visual word within the entire dictionary. We designed an iterative ranking method to measure overall-word significance by considering the relationship between all latent topics and words. The words with higher values are considered meaningful with more significant discriminative power in differentiating medical images. We evaluated our method on two public medical imaging datasets and it showed improved retrieval accuracy and efficiency.

Enhanced anisotropic effective g factors of an Al0.25Ga0.75N/GaN heterostructure based quantum point contact.

Gate-defined quantum point contacts (QPCs) were fabricated with Al0.25Ga0.75N/GaN heterostructures grown by metal-organic chemical vapor deposition (MOCVD). In the transport study of the Zeeman effect, greatly enhanced effective g factors (g*) were obtained. The in-plane g* is found to be 5.5 ± 0.6, 4.8 ± 0.4, and 4.2 ± 0.4 for the first to the third subband, respectively. Similarly, the out-of-plane g* is 8.3 ± 0.6, 6.7 ± 0.7, and 5.1 ± 0.7. Increasing g* with the population of odd-numbered spin-splitted subbands are obtained at 14 T. This portion of increase is assumed to arise from the exchange interaction in one-dimensional systems. A careful analysis shows that not only the exchange interaction but the spin-orbit interaction (SOI) in the strongly confined QPC contributes to the enhancement and anisotropy of g* in different subbands. An approach to distinguish the respective contributions from the SOI and exchange interaction is therefore proposed.

Analyzing Eye Paths Using Fractals.

Visual patterns reflect the anatomical and cognitive background underlying process governing how we perceive information, influenced by stimulus characteristics and our own visual perception. These patterns are both spatially complex and display self-similarity seen in fractal geometry at different scales, making them challenging to measure using the traditional topological dimensions used in Euclidean geometry.However, methods for measuring eye gaze patterns using fractals have shown success in quantifying geometric complexity, matchability, and implementation into machine learning methods. This success is due to the inherent capabilities that fractals possess when reducing dimensionality using Hilbert curves, measuring temporal complexity using the Higuchi fractal dimension (HFD), and determining geometric complexity using the Minkowski-Bouligand dimension.Understanding the many applications of fractals when measuring and analyzing eye gaze patterns can extend the current growing body of knowledge by identifying markers tied to neurological pathology. Additionally, in future work, fractals can facilitate defining imaging modalities in eye tracking diagnostics by exploiting their capability to acquire multiscale information, including complementary functions, structures, and dynamics.

Use of Artificial Intelligence in the Prediction of Chiari Malformation Type 1 Recurrence After Posterior Fossa Decompressive Surgery.

Purpose The purpose of this study was to train a deep learning-based method for the prediction of postoperative recurrence of symptoms in Chiari malformation type 1 (CM1) patients undergoing surgery. Studies suggest that certain radiological and clinical features do exist in patients with treatment failure, though these are inconsistent and poorly defined. Methodology This study was a retrospective cohort study of patients who underwent primary surgical intervention for CM1 from January 2010 to May 2020. Only patients who completed pre- and postoperative 12-item short form (SF-12) surveys were included and these were used to classify the recurrence or persistence of symptoms. Forty patients had an improvement in overall symptoms while 17 had recurrence or persistence. After magnetic resonance imaging (MRI) data augmentation, a ResNet50, pre-trained on the ImageNet dataset, was used for feature extraction, and then clustering-constrained attention multiple instance learning (CLAM), a weakly supervised multi-instance learning framework, was trained for prediction of recurrence. Five-fold cross-validation was used for the development of MRI only, clinical features only, and a combined machine learning model. Results This study included 57 patients who underwent CM1 decompression. The recurrence rate was 30%. The combined model incorporating MRI, pre-operative SF-12 physical component scale (PCS), and extent of cerebellar ectopia performed best with an area under the curve (AUC) of 0.71 and an F1 score of 0.74. Conclusion This is the first study to our knowledge to explore the prediction of postoperative recurrence of symptoms in CM1 patients using machine learning methods and represents the first step toward developing a clinically useful prognostication machine learning model. Further studies utilizing a similar deep learning approach with a larger sample size are needed to improve the performance.

A Deep Learning Framework for Skull Stripping in Brain MRI.

Skull-stripping, an important pre-processing step in neuroimage computing, involves the automated removal of non-brain anatomy (such as the skull, eyes, and ears) from brain images to facilitate brain segmentation and analysis. Manual segmentation is still practiced, but it is time-consuming and highly dependent on the expertise of clinicians or image analysts. Prior studies have developed various skull-stripping algorithms that perform well on brains with mild or no structural abnormalities. Nonetheless, they were not able to address the issue for brains with significant morphological changes, such as those caused by brain tumors, particularly when the tumors are located near the skull's border. In such cases, a portion of the normal brain may be stripped, or the reverse may occur during skull stripping. To address this limitation, we propose to use a novel deep learning framework based on nnUNet for skull stripping in brain MRI. Two publicly available datasets were used to evaluate the proposed method, including a normal brain MRI dataset - The Neurofeedback Skull-stripped Repository (NFBS), and a brain tumor MRI dataset - The Cancer Genome Atlas (TCGA). The method proposed in the study performed better than six other current methods, namely BSE, ROBEX, UNet, SC-UNet, MV-UNet, and 3D U-Net. The proposed method achieved an average Dice coefficient of 0.9960, a sensitivity of 0.9999, and a specificity of 0.9996 on the NFBS dataset, and an average Dice coefficient of 0.9296, a sensitivity of 0.9288, a specificity of 0.9866 and an accuracy of 0.9762 on the TCGA brain tumor dataset.

Radiomics and Machine Learning in Brain Tumors and Their Habitat: A Systematic Review.

Radiomics is a rapidly evolving field that involves extracting and analysing quantitative features from medical images, such as computed tomography or magnetic resonance images. Radiomics has shown promise in brain tumor diagnosis and patient-prognosis prediction by providing more detailed and objective information about tumors' features than can be obtained from the visual inspection of the images alone. Radiomics data can be analyzed to determine their correlation with a tumor's genetic status and grade, as well as in the assessment of its recurrence vs. therapeutic response, among other features. In consideration of the multi-parametric and high-dimensional space of features extracted by radiomics, machine learning can further improve tumor diagnosis, treatment response, and patients' prognoses. There is a growing recognition that tumors and their microenvironments (habitats) mutually influence each other-tumor cells can alter the microenvironment to increase their growth and survival. At the same time, habitats can also influence the behavior of tumor cells. In this systematic review, we investigate the current limitations and future developments in radiomics and machine learning in analysing brain tumors and their habitats.

Artificial Intelligence-Assisted Classification of Gliomas Using Whole Slide Images.

CONTEXT.­: Glioma is the most common primary brain tumor in adults. The diagnosis and grading of different pathological subtypes of glioma is essential in treatment planning and prognosis. OBJECTIVE.­: To propose a deep learning-based approach for the automated classification of glioma histopathology images. Two classification methods, the ensemble method based on 2 binary classifiers and the multiclass method using a single multiclass classifier, were implemented to classify glioma images into astrocytoma, oligodendroglioma, and glioblastoma, according to the 5th edition of the World Health Organization classification of central nervous system tumors, published in 2021. DESIGN.­: We tested 2 different deep neural network architectures (VGG19 and ResNet50) and extensively validated the proposed approach based on The Cancer Genome Atlas data set (n = 700). We also studied the effects of stain normalization and data augmentation on the glioma classification task. RESULTS.­: With the binary classifiers, our model could distinguish astrocytoma and oligodendroglioma (combined) from glioblastoma with an accuracy of 0.917 (area under the curve [AUC] = 0.976) and astrocytoma from oligodendroglioma (accuracy = 0.821, AUC = 0.865). The multiclass method (accuracy = 0.861, AUC = 0.961) outperformed the ensemble method (accuracy = 0.847, AUC = 0.933) with the best performance displayed by the ResNet50 architecture. CONCLUSIONS.­: With the high performance of our model (>80%), the proposed method can assist pathologists and physicians to support examination and differential diagnosis of glioma histopathology images, with the aim to expedite personalized medical care.

Domain-Specific Pre-training Improves Confidence in Whole Slide Image Classification.

Whole Slide Images (WSIs) or histopathology images are used in digital pathology. WSIs pose great challenges to deep learning models for clinical diagnosis, owing to their size and lack of pixel-level annotations. With the recent advancements in computational pathology, newer multiple-instance learning-based models have been proposed. Multiple-instance learning for WSIs necessitates creating patches and uses the encoding of these patches for diagnosis. These models use generic pre-trained models (ResNet-50 pre-trained on ImageNet) for patch encoding. The recently proposed KimiaNet, a DenseNet121 model pre-trained on TCGA slides, is a domain-specific pre-trained model. This paper shows the effect of domain-specific pre-training on WSI classification. To investigate the effect of domain-specific pre-training, we considered the current state-of-the-art multiple-instance learning models, 1) CLAM, an attention-based model, and 2) TransMIL, a self-attention-based model, and evaluated the models' confidence and predictive performance in detecting primary brain tumors - gliomas. Domain-specific pre-training improves the confidence of the models and also achieves a new state-of-the-art performance of WSI-based glioma subtype classification, showing a high clinical applicability in assisting glioma diagnosis. We will publicly share our code and experimental results at https://github.com/soham-chitnis10/WSI-domain-specific.