Development and Evaluation of a Natural Language Processing Annotation Tool to Facilitate Phenotyping of Cognitive Status in Electronic Health Records: Diagnostic Study.

BACKGROUND: Electronic health records (EHRs) with large sample sizes and rich information offer great potential for dementia research, but current methods of phenotyping cognitive status are not scalable. OBJECTIVE: The aim of this study was to evaluate whether natural language processing (NLP)-powered semiautomated annotation can improve the speed and interrater reliability of chart reviews for phenotyping cognitive status. METHODS: In this diagnostic study, we developed and evaluated a semiautomated NLP-powered annotation tool (NAT) to facilitate phenotyping of cognitive status. Clinical experts adjudicated the cognitive status of 627 patients at Mass General Brigham (MGB) health care, using NAT or traditional chart reviews. Patient charts contained EHR data from two data sets: (1) records from January 1, 2017, to December 31, 2018, for 100 Medicare beneficiaries from the MGB Accountable Care Organization and (2) records from 2 years prior to COVID-19 diagnosis to the date of COVID-19 diagnosis for 527 MGB patients. All EHR data from the relevant period were extracted; diagnosis codes, medications, and laboratory test values were processed and summarized; clinical notes were processed through an NLP pipeline; and a web tool was developed to present an integrated view of all data. Cognitive status was rated as cognitively normal, cognitively impaired, or undetermined. Assessment time and interrater agreement of NAT compared to manual chart reviews for cognitive status phenotyping was evaluated. RESULTS: NAT adjudication provided higher interrater agreement (Cohen κ=0.89 vs κ=0.80) and significant speed up (time difference mean 1.4, SD 1.3 minutes; P<.001; ratio median 2.2, min-max 0.4-20) over manual chart reviews. There was moderate agreement with manual chart reviews (Cohen κ=0.67). In the cases that exhibited disagreement with manual chart reviews, NAT adjudication was able to produce assessments that had broader clinical consensus due to its integrated view of highlighted relevant information and semiautomated NLP features. CONCLUSIONS: NAT adjudication improves the speed and interrater reliability for phenotyping cognitive status compared to manual chart reviews. This study underscores the potential of an NLP-based clinically adjudicated method to build large-scale dementia research cohorts from EHRs.

Preferential rabbit antibody responses to C-termini of NOTCH3 peptide immunogens.

Antibodies raised in peptide-immunized rabbits have been used in biological research for decades. Although there has been wide implementation of this approach, specific proteins are occasionally difficult to target for multiple reasons. One consideration that was noted in mice is that humoral responses may preferentially target the carboxyl terminus of the peptide sequence which is not present in the intact protein. To shed light on the frequency of preferential rabbit antibody responses to C-termini of peptide immunogens, we present our experience with generation of rabbit antibodies to human NOTCH3. A total of 23 antibodies were raised against 10 peptide sequences of human NOTCH3. Over 70% (16 of 23) of these polyclonal antibodies were determined to be C-terminal preferring: NOTCH3 peptide-reactive antibodies largely targeted the terminating free carboxyl group of the immunizing peptide. The antibodies that preferred C-terminal epitopes reacted weakly or not at all with recombinant target sequences with extension the C-terminus that eliminated the free carboxyl group of the immunogen structure; furthermore, each of these antisera revealed no antibody reactivity to proteins truncated before the C-terminus of the immunogen. In immunocytochemical applications of these anti-peptide antibodies, we similarly found reactivity to recombinant targets that best binding to cells expressing the free C-terminus of the immunizing sequence. In aggregate, our experience demonstrates a strong propensity for rabbits to mount antibody responses to C-terminal epitopes of NOTCH3-derived peptides which is predicted to limit their use against the native protein. We discuss some potential approaches to overcome this bias that could improve the efficiency of generation of antibodies in this commonly utilized experimental paradigm.

Hierarchical discriminative learning improves visual representations of biomedical microscopy.

Learning high-quality, self-supervised, visual representations is essential to advance the role of computer vision in biomedical microscopy and clinical medicine. Previous work has focused on self-supervised representation learning (SSL) methods developed for instance discrimination and applied them directly to image patches, or fields-of-view, sampled from gigapixel whole-slide images (WSIs) used for cancer diagnosis. However, this strategy is limited because it (1) assumes patches from the same patient are independent, (2) neglects the patient-slide-patch hierarchy of clinical biomedical microscopy, and (3) requires strong data augmentations that can degrade downstream performance. Importantly, sampled patches from WSIs of a patient's tumor are a diverse set of image examples that capture the same underlying cancer diagnosis. This motivated HiDisc, a data-driven method that leverages the inherent patient-slide-patch hierarchy of clinical biomedical microscopy to define a hierarchical discriminative learning task that implicitly learns features of the underlying diagnosis. HiDisc uses a self-supervised contrastive learning framework in which positive patch pairs are defined based on a common ancestry in the data hierarchy, and a unified patch, slide, and patient discriminative learning objective is used for visual SSL. We benchmark HiDisc visual representations on two vision tasks using two biomedical microscopy datasets, and demonstrate that (1) HiDisc pretraining outperforms current state-of-the-art self-supervised pretraining methods for cancer diagnosis and genetic mutation prediction, and (2) HiDisc learns high-quality visual representations using natural patch diversity without strong data augmentations.

Concentration of non-myocyte proteins in arterial media of cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy.

The most common inherited cause of vascular dementia and stroke, cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), is caused by mutations in NOTCH3. Post-translationally altered NOTCH3 accumulates in the vascular media of CADASIL arteries in areas of the vessels that exhibit profound cellular degeneration. The identification of molecules that concentrate in the same location as pathological NOTCH3 may shed light on processes that drive cytopathology in CADASIL. We performed a two phase immunohistochemical screen of markers identified in the Human Protein Atlas to identify new proteins that accumulate in the vascular media in a pattern similar to pathological NOTCH3. In phase one, none of 16 smooth muscle cell (SMC) localized antigens exhibited NOTCH3-like patterns of expression; however, several exhibited disease-dependent patterns of expression, with antibodies directed against FAM124A, GZMM, MTFR1, and ST6GAL demonstrating higher expression in controls than CADASIL. In contrast, in phase two of the study that included 56 non-SMC markers, two proteins, CD63 and CTSH, localized to the same regions as pathological NOTCH3, which was verified by VesSeg, a customized algorithm that assigns relative location of antigens within the layers of the vessel. Proximity ligation assays support complex formation between NOTCH3 fragments and CD63 in degenerating CADASIL media. Interestingly, in normal mouse brain, the two novel CADASIL markers, CD63 and CTSH, are expressed in non-SMC vascular cells. The identification of new proteins that concentrate in CADASIL vascular media demonstrates the utility of querying publicly available protein databases in specific neurological diseases and uncovers unexpected, non-SMC origins of pathological antigens in small vessel disease.

Artificial-intelligence-based molecular classification of diffuse gliomas using rapid, label-free optical imaging.

Molecular classification has transformed the management of brain tumors by enabling more accurate prognostication and personalized treatment. However, timely molecular diagnostic testing for patients with brain tumors is limited, complicating surgical and adjuvant treatment and obstructing clinical trial enrollment. In this study, we developed DeepGlioma, a rapid (<90 seconds), artificial-intelligence-based diagnostic screening system to streamline the molecular diagnosis of diffuse gliomas. DeepGlioma is trained using a multimodal dataset that includes stimulated Raman histology (SRH); a rapid, label-free, non-consumptive, optical imaging method; and large-scale, public genomic data. In a prospective, multicenter, international testing cohort of patients with diffuse glioma (n = 153) who underwent real-time SRH imaging, we demonstrate that DeepGlioma can predict the molecular alterations used by the World Health Organization to define the adult-type diffuse glioma taxonomy (IDH mutation, 1p19q co-deletion and ATRX mutation), achieving a mean molecular classification accuracy of 93.3 ± 1.6%. Our results represent how artificial intelligence and optical histology can be used to provide a rapid and scalable adjunct to wet lab methods for the molecular screening of patients with diffuse glioma.

102 AI-Based Molecular Classification of Diffuse Gliomas using Rapid, Label-Free Optical Imaging.

INTRODUCTION: Molecular classification has transformed the management of brain tumors by enabling more accurate prognostication and personalized treatment. Access to timely molecular diagnostic testing for brain tumor patients is limited, complicating surgical and adjuvant treatment and obstructing clinical trial enrollment. METHODS: By combining stimulated Raman histology (SRH), a rapid, label-free, non-consumptive, optical imaging method, and deep learning-based image classification, we are able to predict the molecular genetic features used by the World Health Organization (WHO) to define the adult-type diffuse glioma taxonomy, including IDH-1/2, 1p19q-codeletion, and ATRX loss. We developed a multimodal deep neural network training strategy that uses both SRH images and large-scale, public diffuse glioma genomic data (i.e. TCGA, CGGA, etc.) in order to achieve optimal molecular classification performance. RESULTS: One institution was used for model training (University of Michigan) and four institutions (NYU, UCSF, Medical University of Vienna, and University Hospital Cologne) were included for patient enrollment in the prospective testing cohort. Using our system, called DeepGlioma, we achieved an average molecular genetic classification accuracy of 93.2% and identified the correct diffuse glioma molecular subgroup with 91.5% accuracy within 2 minutes in the operating room. DeepGlioma outperformed conventional IDH1-R132H immunohistochemistry (94.2% versus 91.4% accuracy) as a first-line molecular diagnostic screening method for diffuse gliomas and can detect canonical and non-canonical IDH mutations. CONCLUSIONS: Our results demonstrate how artificial intelligence and optical histology can be used to provide a rapid and scalable alternative to wet lab methods for the molecular diagnosis of brain tumor patients during surgery.

OpenSRH: optimizing brain tumor surgery using intraoperative stimulated Raman histology.

Accurate intraoperative diagnosis is essential for providing safe and effective care during brain tumor surgery. Our standard-of-care diagnostic methods are time, resource, and labor intensive, which restricts access to optimal surgical treatments. To address these limitations, we propose an alternative workflow that combines stimulated Raman histology (SRH), a rapid optical imaging method, with deep learning-based automated interpretation of SRH images for intraoperative brain tumor diagnosis and real-time surgical decision support. Here, we present OpenSRH, the first public dataset of clinical SRH images from 300+ brain tumors patients and 1300+ unique whole slide optical images. OpenSRH contains data from the most common brain tumors diagnoses, full pathologic annotations, whole slide tumor segmentations, raw and processed optical imaging data for end-to-end model development and validation. We provide a framework for patch-based whole slide SRH classification and inference using weak (i.e. patient-level) diagnostic labels. Finally, we benchmark two computer vision tasks: multiclass histologic brain tumor classification and patch-based contrastive representation learning. We hope OpenSRH will facilitate the clinical translation of rapid optical imaging and real-time ML-based surgical decision support in order to improve the access, safety, and efficacy of cancer surgery in the era of precision medicine. Dataset access, code, and benchmarks are available at https://opensrh.mlins.org.

Rapid Automated Analysis of Skull Base Tumor Specimens Using Intraoperative Optical Imaging and Artificial Intelligence.

BACKGROUND: Accurate specimen analysis of skull base tumors is essential for providing personalized surgical treatment strategies. Intraoperative specimen interpretation can be challenging because of the wide range of skull base pathologies and lack of intraoperative pathology resources. OBJECTIVE: To develop an independent and parallel intraoperative workflow that can provide rapid and accurate skull base tumor specimen analysis using label-free optical imaging and artificial intelligence. METHODS: We used a fiber laser-based, label-free, nonconsumptive, high-resolution microscopy method (<60 seconds per 1 × 1 mm2), called stimulated Raman histology (SRH), to image a consecutive, multicenter cohort of patients with skull base tumor. SRH images were then used to train a convolutional neural network model using 3 representation learning strategies: cross-entropy, self-supervised contrastive learning, and supervised contrastive learning. Our trained convolutional neural network models were tested on a held-out, multicenter SRH data set. RESULTS: SRH was able to image the diagnostic features of both benign and malignant skull base tumors. Of the 3 representation learning strategies, supervised contrastive learning most effectively learned the distinctive and diagnostic SRH image features for each of the skull base tumor types. In our multicenter testing set, cross-entropy achieved an overall diagnostic accuracy of 91.5%, self-supervised contrastive learning 83.9%, and supervised contrastive learning 96.6%. Our trained model was able to segment tumor-normal margins and detect regions of microscopic tumor infiltration in meningioma SRH images. CONCLUSION: SRH with trained artificial intelligence models can provide rapid and accurate intraoperative analysis of skull base tumor specimens to inform surgical decision-making.