Intraoperative diagnosis is essential for providing safe and effective care during cancer surgery 1. The existing workflow for intraoperative diagnosis based on hematoxylin and eosin-staining of processed tissue is time-, resource-, and labor-intensive 2,3. Moreover, interpretation of intraoperative histologic images is dependent on a contracting, unevenly distributed pathology workforce 4. Here, we report a parallel workflow that combines stimulated Raman histology (SRH) 5-7 , a label-free optical imaging method, and deep convolutional neural networks (CNN) to predict diagnosis at the bedside in near real-time in an automated fashion. Specifically, our CNN, trained on over 2.5 million SRH images, predicts brain tumor diagnosis in the operating room in under 150 seconds, an order of magnitude faster than conventional techniques (e.g., 20-30 minutes) 2. In a multicenter, prospective clinical trial (n = 278) we demonstrated that CNN-based diagnosis of SRH images was non-inferior to pathologist-based interpretation of conventional histologic images (overall accuracy, 94.6% vs. 93.9%). Our CNN learned a hierarchy of recognizable histologic feature representations to classify the major histopathologic classes of brain tumors. Additionally, we implemented a semantic segmentation method to identify tumor infiltrated, diagnostic regions within SRH images. These results demonstrate how intraoperative cancer diagnosis can be streamlined, creating a complimentary pathway for tissue diagnosis that is independent of a traditional pathology laboratory.

No abstract

OBJECTIVEPituitary adenomas occur in a heterogeneous patient population with diverse perioperative risk factors, endocrinopathies, and other tumor-related comorbidities. This heterogeneity makes predicting postoperative outcomes challenging when using traditional scoring systems. Modern machine learning algorithms can automatically identify the most predictive risk factors and learn complex risk-factor interactions using training data to build a robust predictive model that can generalize to new patient cohorts. The authors sought to build a predictive model using supervised machine learning to accurately predict early outcomes of pituitary adenoma surgery.METHODSA retrospective cohort of 400 consecutive pituitary adenoma patients was used. Patient variables/predictive features were limited to common patient characteristics to improve model implementation. Univariate and multivariate odds ratio analysis was performed to identify individual risk factors for common postoperative complications and to compare risk factors with model predictors. The study population was split into 300 training/validation patients and 100 testing patients to train and evaluate four machine learning models using binary classification accuracy for predicting early outcomes.RESULTSThe study included a total of 400 patients. The mean ± SD patient age was 53.9 ± 16.3 years, 59.8% of patients had nonfunctioning adenomas and 84.7% had macroadenomas, and the mean body mass index (BMI) was 32.6 ± 7.8 (58.0% obesity rate). Multivariate odds ratio analysis demonstrated that age < 40 years was associated with a 2.86 greater odds of postoperative diabetes insipidus and that nonobese patients (BMI < 30) were 2.2 times more likely to develop postoperative hyponatremia. Using broad criteria for a poor early postoperative outcome—major medical and early surgical complications, extended length of stay, emergency department admission, inpatient readmission, and death—31.0% of patients met criteria for a poor early outcome. After model training, a logistic regression model with elastic net (LR-EN) regularization best predicted early postoperative outcomes of pituitary adenoma surgery on the 100-patient testing set—sensitivity 68.0%, specificity 93.3%, overall accuracy 87.0%. The receiver operating characteristic and precision-recall curves for the LR-EN model had areas under the curve of 82.7 and 69.5, respectively. The most important predictive variables were lowest perioperative sodium, age, BMI, highest perioperative sodium, and Cushing’s disease.CONCLUSIONSEarly postoperative outcomes of pituitary adenoma surgery can be predicted with 87% accuracy using a machine learning approach. These results provide insight into how predictive modeling using machine learning can be used to improve the perioperative management of pituitary adenoma patients.

High-risk HPV (hrHPV) is the leading etiologic factor in oropharyngeal cancer. HPV-positive oropharyngeal tumors generally respond well to therapy, with complete recovery in approximately 80% of patients. However, it remains unclear why some patients are non-responsive to treatment, with 20% of patients recurring within 5 years. In this study, viral factors were examined for possible clues to differences in tumor behavior. Oropharynx tumors that responded well to therapy were compared to those that persisted and recurred. Viral oncogene alternate transcripts were assessed and cellular sites of viral integration were mapped and sequenced. Effects of integration on gene expression were assessed by transcript analysis at the integration sites. All of the tumors demonstrated active viral oncogenesis, indicated by expression of HPV E6 and E7 oncogenes and alternate E6 splicing. In the responsive tumors, HPV integration occurred exclusively in intergenic chromosome regions, except for one tumor with viral integration into TP63. Each recurrent tumor exhibited complex HPV integration patterns into cancer-associated genes, including: TNFRSF13B, SCN2A, SH2B1, UBE2V2, SMOC1, NFIA, and SEMA6D. Disrupted cellular transcripts were identified in the region of integration in four of the seven affected genes. Implications Integration of transcriptionally active hrHPV into cellular intergenic regions associates with tumor behavior by altering gene expression.

J Neurosurg 124:811-816, 2016P atients suffering from pathologies of the ventral and ventrolateral skull base often require surgical intervention. Access to the skull base traditionally required extensive tissue manipulation, but recent advances in endoscopic techniques have allowed access to the skull base using less destructive techniques via the nostril as a natural corridor. The endoscopic endonasal approach (EEA) is employed in the treatment of pituitary adenomas, Rathke's cleft cysts, ventral cranial base meningiomas, craniopharyngiomas, chordomas, olfactory neuroblastomas, and sinonasal carcinomas involving the skull base, among other pathologies. Exposure and resection often require drilling of bone adjacent to critical neurovascular structures. Cranial nerve damage can result in catastrophic disability, and vascular injury may be maiming or fatal. 6 Although residents are trained to perform the EEA by participating in operations on patients while under close supervision, critical drilling is often still done by the attending surgeon. Endoscopic endonasal drilling is somewhat unique in regard to the instruments used, the long reach required, and the restricted angulation. Cadaver training is a viable option for teaching such skills, when available. Sensory feedback from cadaver bone remains largely unchanged from live tissue, and the anatomy remains intact. However, cadavers are expensive and require specialized equipment, staff, and facilities. Availability of specimens or facilities, in most cases, precludes the use of cadavers for basic technique training; rather, their value is maximized when used by more experienced operators. Alternatively, a number of virtual reality (VR) simulators are available for EEA, but they often lack realistic drilling haptics, and the initial costs could be high. 2,10We successfully developed a physical simulator for ventriculostomy placement and demonstrated its validity. 15 Based on that experience, we also sought to create an inexpensive, high-fidelity simulator by which the drilling aspects of the EEA could be practiced by trainees once they advance to the point of surgical participation, thereby reducing the risk to patients. This endeavor, a collaborative effort of neurosurgeons, otolaryngologists, engineers, and an education/research specialist, is described in terms of simulator design, training setup, and validation process. In this paper, the authors present a physical model developed to teach surgeons the requisite drilling techniques when using an endoscopic endonasal approach (EEA) to the skull base. EEA is increasingly used for treating pathologies of the ventral and ventrolateral cranial base. Endonasal drilling is a unique skill in terms of the instruments used, the long reach required, and the restricted angulation, and gaining competency requires much practice. Based on the successful experience in creating custom simulators, the authors used 3D printing to build an EEA training model from post-processed thin-cut head CT scans, formulating the materials ...