Ultra-fast deep-learned CNS tumour classification during surgery.

Central nervous system tumours represent one of the most lethal cancer types, particularly among children1. Primary treatment includes neurosurgical resection of the tumour, in which a delicate balance must be struck between maximizing the extent of resection and minimizing risk of neurological damage and comorbidity2,3. However, surgeons have limited knowledge of the precise tumour type prior to surgery. Current standard practice relies on preoperative imaging and intraoperative histological analysis, but these are not always conclusive and occasionally wrong. Using rapid nanopore sequencing, a sparse methylation profile can be obtained during surgery4. Here we developed Sturgeon, a patient-agnostic transfer-learned neural network, to enable molecular subclassification of central nervous system tumours based on such sparse profiles. Sturgeon delivered an accurate diagnosis within 40 minutes after starting sequencing in 45 out of 50 retrospectively sequenced samples (abstaining from diagnosis of the other 5 samples). Furthermore, we demonstrated its applicability in real time during 25 surgeries, achieving a diagnostic turnaround time of less than 90 min. Of these, 18 (72%) diagnoses were correct and 7 did not reach the required confidence threshold. We conclude that machine-learned diagnosis based on low-cost intraoperative sequencing can assist neurosurgical decision-making, potentially preventing neurological comorbidity and avoiding additional surgeries.

Inter-rater agreement in glioma segmentations on longitudinal MRI.

BACKGROUND: Tumor segmentation of glioma on MRI is a technique to monitor, quantify and report disease progression. Manual MRI segmentation is the gold standard but very labor intensive. At present the quality of this gold standard is not known for different stages of the disease, and prior work has mainly focused on treatment-naive glioblastoma. In this paper we studied the inter-rater agreement of manual MRI segmentation of glioblastoma and WHO grade II-III glioma for novices and experts at three stages of disease. We also studied the impact of inter-observer variation on extent of resection and growth rate. METHODS: In 20 patients with WHO grade IV glioblastoma and 20 patients with WHO grade II-III glioma (defined as non-glioblastoma) both the enhancing and non-enhancing tumor elements were segmented on MRI, using specialized software, by four novices and four experts before surgery, after surgery and at time of tumor progression. We used the generalized conformity index (GCI) and the intra-class correlation coefficient (ICC) of tumor volume as main outcome measures for inter-rater agreement. RESULTS: For glioblastoma, segmentations by experts and novices were comparable. The inter-rater agreement of enhancing tumor elements was excellent before surgery (GCI 0.79, ICC 0.99) poor after surgery (GCI 0.32, ICC 0.92), and good at progression (GCI 0.65, ICC 0.91). For non-glioblastoma, the inter-rater agreement was generally higher between experts than between novices. The inter-rater agreement was excellent between experts before surgery (GCI 0.77, ICC 0.92), was reasonable after surgery (GCI 0.48, ICC 0.84), and good at progression (GCI 0.60, ICC 0.80). The inter-rater agreement was good between novices before surgery (GCI 0.66, ICC 0.73), was poor after surgery (GCI 0.33, ICC 0.55), and poor at progression (GCI 0.36, ICC 0.73). Further analysis showed that the lower inter-rater agreement of segmentation on postoperative MRI could only partly be explained by the smaller volumes and fragmentation of residual tumor. The median interquartile range of extent of resection between raters was 8.3% and of growth rate was 0.22 mm/year. CONCLUSION: Manual tumor segmentations on MRI have reasonable agreement for use in spatial and volumetric analysis. Agreement in spatial overlap is of concern with segmentation after surgery for glioblastoma and with segmentation of non-glioblastoma by non-experts.

Diagnostic Accuracy of Neuroimaging to Delineate Diffuse Gliomas within the Brain: A Meta-Analysis.

BACKGROUND: Brain imaging in diffuse glioma is used for diagnosis, treatment planning, and follow-up. PURPOSE: In this meta-analysis, we address the diagnostic accuracy of imaging to delineate diffuse glioma. DATA SOURCES: We systematically searched studies of adults with diffuse gliomas and correlation of imaging with histopathology. STUDY SELECTION: Study inclusion was based on quality criteria. Individual patient data were used, if available. DATA ANALYSIS: A hierarchic summary receiver operating characteristic method was applied. Low- and high-grade gliomas were analyzed in subgroups. DATA SYNTHESIS: Sixty-one studies described 3532 samples in 1309 patients. The mean Standard for Reporting of Diagnostic Accuracy score (13/25) indicated suboptimal reporting quality. For diffuse gliomas as a whole, the diagnostic accuracy was best with T2-weighted imaging, measured as area under the curve, false-positive rate, true-positive rate, and diagnostic odds ratio of 95.6%, 3.3%, 82%, and 152. For low-grade gliomas, the diagnostic accuracy of T2-weighted imaging as a reference was 89.0%, 0.4%, 44.7%, and 205; and for high-grade gliomas, with T1-weighted gadolinium-enhanced MR imaging as a reference, it was 80.7%, 16.8%, 73.3%, and 14.8. In high-grade gliomas, MR spectroscopy (85.7%, 35.0%, 85.7%, and 12.4) and 11C methionine-PET (85.1%, 38.7%, 93.7%, and 26.6) performed better than the reference imaging. LIMITATIONS: True-negative samples were underrepresented in these data, so false-positive rates are probably less reliable than true-positive rates. Multimodality imaging data were unavailable. CONCLUSIONS: The diagnostic accuracy of commonly used imaging is better for delineation of low-grade gliomas than high-grade gliomas on the basis of limited evidence. Improvement is indicated from advanced techniques, such as MR spectroscopy and PET.

Effects of physician-based emergency medical service dispatch in severe traumatic brain injury on prehospital run time.

INTRODUCTION: Prehospital care by physician-based helicopter emergency medical services (P-HEMS) may prolong total prehospital run time. This has raised an issue of debate about the benefits of these services in traumatic brain injury (TBI). We therefore investigated the effects of P-HEMS dispatch on prehospital run time and outcome in severe TBI. METHODS: Prehospital run times of 497 patients with severe TBI who were solely treated by a paramedic EMS (n = 125) or an EMS/P-HEMS combination (n = 372) were retrospectively analyzed. Other study parameters included the injury severity score (ISS), Glasgow Coma Scale (GCS), prehospital endotracheal intubation and predicted and observed outcome rates. RESULTS: Patients who received P-HEMS care were younger and had higher ISS values than solely EMS-treated patients (10%; P = 0.04). The overall prehospital run time was 74 ± 54 min, with similar out-of-hospital times for EMS and P-HEMS treated patients. Prehospital endotracheal intubation was more frequently performed in the P-HEMS group (88%) than in the EMS group (35%; P<0.001). The prehospital run time for intubated patients was similar for P-HEMS (66 (51-80)min) and EMS-treated patients (59 (41-88 min). Unexpectedly, mortality probability scores and observed outcome scores were less favourable for EMS-treated patients when compared to patients treated by P-HEMS. CONCLUSION: P-HEMS dispatch does not increase prehospital run times in severe TBI, while it assures prehospital intubation of TBI patients by a well-trained physician. Our data however suggest that a subgroup of the most severely injured patients received prehospital care by an EMS, while international guidelines recommend advanced life support by a physician-based EMS in these cases.