Precision of preoperative diagnosis in patients with brain tumor - A prospective study based on "top three list" of differential diagnosis for 1061 patients

An accessible deep learning tool for voxel-wise classification of brain malignancies from perfusion MRI

Noninvasive differential diagnosis of brain tumors is currently based on the assessment of magnetic resonance imaging (MRI) coupled with dynamic susceptibility contrast (DSC). However, a definitive diagnosis often requires neurosurgical interventions that compromise patients' quality of life. We apply deep learning on DSC images from histology-confirmed patients with glioblastoma, metastasis, or lymphoma. The convolutional neural network trained on ∼50,000 voxels from 40 patients provides intratumor probability maps that yield clinical-grade diagnosis. Performance is tested in 400 additional cases and an external validation cohort of 128 patients. The tool reaches a three-way accuracy of 0.78, superior to the conventional MRI metrics cerebral blood volume (0.55) and percentage of signal recovery (0.59), showing high value as a support diagnostic tool. Our open-access software, Diagnosis In Susceptibility Contrast Enhancing Regions for Neuro-oncology (DISCERN), demonstrates its potential in aiding medical decisions for brain tumor diagnosis using standard-of-care MRI.

Advances in the Use of Deep Learning for the Analysis of Magnetic Resonance Image in Neuro-Oncology

Machine Learning is entering a phase of maturity, but its medical applications still lag behind in terms of practical use. The field of oncological radiology (and neuro-oncology in particular) is at the forefront of these developments, now boosted by the success of Deep-Learning methods for the analysis of medical images. This paper reviews in detail some of the most recent advances in the use of Deep Learning in this field, from the broader topic of the development of Machine-Learning-based analytical pipelines to specific instantiations of the use of Deep Learning in neuro-oncology; the latter including its use in the groundbreaking field of ultra-low field magnetic resonance imaging.

Cerebrospinal fluid methylome-based liquid biopsies for accurate malignant brain neoplasm classification

BACKGROUND

Resolving the differential diagnosis between brain metastases (BM), glioblastomas (GBM), and central nervous system lymphomas (CNSL) is an important dilemma for the clinical management of the main three intra-axial brain tumor types. Currently, treatment decisions require invasive diagnostic surgical biopsies that carry risks and morbidity. This study aimed to utilize methylomes from cerebrospinal fluid (CSF), a biofluid proximal to brain tumors, for reliable non-invasive classification that addresses limitations associated with low target abundance in existing approaches.

METHODS

Binomial GLMnet classifiers of tumor type were built, in fifty iterations of 80% discovery sets, using CSF methylomes obtained from 57 BM, GBM, CNSL, and non-neoplastic control patients. Publicly-available tissue methylation profiles (N = 197) on these entities and normal brain parenchyma were used for validation and model optimization.

RESULTS

Models reliably distinguished between BM (area under receiver operating characteristic curve [AUROC] = 0.93, 95% confidence interval [CI]: 0.71-1.0), GBM (AUROC = 0.83, 95% CI: 0.63-1.0), and CNSL (AUROC = 0.91, 95% CI: 0.66-1.0) in independent 20% validation sets. For validation, CSF-based methylome signatures reliably distinguished between tumor types within external tissue samples and tumors from non-neoplastic controls in CSF and tissue. CSF methylome signals were observed to align closely with tissue signatures for each entity. An additional set of optimized CSF-based models, built using tumor-specific features present in tissue data, showed enhanced classification accuracy.

CONCLUSIONS

CSF methylomes are reliable for liquid biopsy-based classification of the major three malignant brain tumor types. We discuss how liquid biopsies may impact brain cancer management in the future by avoiding surgical risks, classifying unbiopsiable tumors, and guiding surgical planning when resection is indicated.

Automatic brain tumour diagnostic method based on a back propagation neural network and an extended set-membership filter

BACKGROUND

Diagnosing brain tumours remains a challenging task in clinical practice. Despite their questionable accuracy, magnetic resonance image (MRI) scans are presently considered the optimal facility for assessing the growth of tumours. However, the efficiency of manual diagnosis is low, and high computational cost and poor convergence restrict the application of machine learning methods. This study aims to design a method that can reliably diagnose brain tumours from MRI scans.

METHODS

First, image pre-processing (which includes background removal, size standardization, noise removal, and contrast enhancement) is utilized to normalize the images. Then, grey level co-occurrence matrix features are selected as texture features of the brain MRI scans. Finally, a method combining a back propagation neural network (BPNN) and an extended set-membership filter (ESMF) is proposed to classify features and perform image classification.

RESULTS

A total of 304 patient MRI series (247 images of brains with tumours and 57 images of normal brains) were included and assessed in this study. The results revealed that our proposed method can achieve an accuracy of 95.40% and has classification accuracies of 97.14% and 88.24% for brain tumour and normal brain, respectively.

CONCLUSION

This study proposes an automatic brain tumour detection model constructed using a combination of BPNN and ESMF. The model is found to be able to accurately classify brain MRI scans as normal or tumour images.

Preoperative Prediction Nomogram Based on Integrated Profiling for Glioblastoma Multiforme in Glioma Patients

Introduction: Traditional classification that divided gliomas into glioblastoma multiformes (GBM) and lower grade gliomas (LGG) based on pathological morphology has been challenged over the past decade by improvements in molecular stratification, however, the reproducibility and diagnostic accuracy of glioma classification still remains poor. This study aimed to establish and validate a novel nomogram for the preoperative diagnosis of GBM by using integrated data combined with feasible baseline characteristics and preoperative tests. Material and method: The models were established in a primary cohort that included 259 glioma patients who had undergone surgical resection and were pathologically diagnosed from March 2014 to May 2016 in the First Affiliated Hospital of Xi'an Jiaotong University. The preoperative data were used to construct three models by the best subset regression, the forward stepwise regression, and the least absolute shrinkage and selection operator, and to furthermore establish the nomogram among those models. The assessment of nomogram was carried out by the discrimination and calibration in internal cohorts and external cohorts. Results and discussion: Out of all three models, model 2 contained eight clinical-related variables, which exhibited the minimum Akaike Information Criterion (173.71) and maximum concordance index (0.894). Compared with the other two models, the integrated discrimination index for model 2 was significantly improved, indicating that the nomogram obtained from model 2 was the most appropriate model. Likewise, the nomogram showed great calibration and significant clinical benefit according to calibration curves and the decision curve analysis. Conclusion: In conclusion, our study showed a novel preoperative model that incorporated clinically relevant variables and imaging features with laboratory data that could be used for preoperative prediction in glioma patients, thus providing more reliable evidence for surgical decision-making.