A robust hybrid fusion segmentation approach for automated tumor diagnosis and classification in brain MR images

In today’s world, Brain Tumor diagnosis plays a significant role in the field of Oncology. The earlier identification of brain tumors increases the compatibility of treatment of patients and offers an efficient diagnostic recommendation from medical practitioners. Nevertheless, accurate segmentation and feature extraction are the vital challenges in brain tumor diagnosis where the handling of higher resolution images increases the processing time of existing classifiers. In this paper, a new robust weighted hybrid fusion classifier has been proposed to identify and classify the tumefaction in the brain which is of the hybridized form of SVM, NB, and KNN (SNK) classifiers. Primarily, the proposed methodology initiates the preprocessing technique such as adaptive fuzzy filtration and skull stripping in order to remove the noises as well as unwanted regions. Subsequently, an automated hybrid segmentation strategy can be carried out to acquire the initial segmentation results, and then their outcomes are compiled together using fusion rules to accurately localize the tumor region. Finally, a Hybrid SNK classifier is implemented in the proposed methodology for categorizing the type of tumefaction in the brain. The hybrid classifier has been compared with the existing state-of-the-art classifier which shows a higher accuracy result of 99.18% while distinguishing the benign and malignant tumors from brain Magnetic Resonance (MR) images.

FWNNet: Presentation of a New Classifier of Brain Tumor Diagnosis Based on Fuzzy Logic and the Wavelet-Based Neural Network Using Machine-Learning Methods

In this paper, we present a novel classifier based on fuzzy logic and wavelet transformation in the form of a neural network. This classifier includes a layer to predict the numerical feature corresponded to labels or classes. The presented classifier is implemented in brain tumor diagnosis. For feature extraction, a fractal model with four Gaussian functions is used. The classification is performed on 2000 MRI images. Regarding the results, the accuracy of the DT, KNN, LDA, NB, MLP, and SVM is 93.5%, 87.6%, 61.5%, 57.5%, 68.5%, and 43.6%, respectively. Based on the results, the presented FWNNet illustrates the highest accuracy of 100% with the fractal feature extraction method and brain tumor diagnosis based on MRI images. Based on the results, the best classifier for diagnosis of the brain tumor is FWNNet architecture. However, the second and third high-performance classifiers are the DT and KNN, respectively. Moreover, the presented FWNNet method is implemented for the segmentation of brain tumors. In this paper, we present a novel supervised segmentation method based on the FWNNet layer. In the training process, input images with a sweeping filter should be reshaped to vectors that correspond to reshaped ground truth images. In the training process, we performed a PSO algorithm to optimize the gradient descent algorithm. For this purpose, 80 MRI images are used to segment the brain tumor. Based on the results of the ROC curve, it can be estimated that the presented layer can segment the brain tumor with a high true-positive rate.

Doctor's Dilemma: Evaluating an Explainable Subtractive Spatial Lightweight Convolutional Neural Network for Brain Tumor Diagnosis

In Medicine Deep Learning has become an essential tool to achieve outstanding diagnosis on image data. However, one critical problem is that Deep Learning comes with complicated, black-box models so it is not possible to analyze their trust level directly. So, Explainable Artificial Intelligence (XAI) methods are used to build additional interfaces for explaining how the model has reached the outputs by moving from the input data. Of course, that's again another competitive problem to analyze if such methods are successful according to the human view. So, this paper comes with two important research efforts: (1) to build an explainable deep learning model targeting medical image analysis, and (2) to evaluate the trust level of this model via several evaluation works including human contribution. The target problem was selected as the brain tumor classification, which is a remarkable, competitive medical image-based problem for Deep Learning. In the study, MR-based pre-processed brain images were received by the Subtractive Spatial Lightweight Convolutional Neural Network (SSLW-CNN) model, which includes additional operators to reduce the complexity of classification. In order to ensure the explainable background, the model also included Class Activation Mapping (CAM). It is important to evaluate the trust level of a successful model. So, numerical success rates of the SSLW-CNN were evaluated based on the peak signal-to-noise ratio (PSNR), computational time, computational overhead, and brain tumor classification accuracy. The objective of the proposed SSLW-CNN model is to obtain faster and good tumor classification with lesser time. The results illustrate that the SSLW-CNN model provides better performance of PSNR which is enhanced by 8%, classification accuracy is improved by 33%, computation time is reduced by 19%, computation overhead is decreased by 23%, and classification time is minimized by 13%, as compared to state-of-the-art works. Because the model provided good numerical results, it was then evaluated in terms of XAI perspective by including doctor-model based evaluations such as feedback CAM visualizations, usability, expert surveys, comparisons of CAM with other XAI methods, and manual diagnosis comparison. The results show that the SSLW-CNN provides good performance on brain tumor diagnosis and ensures a trustworthy solution for the doctors.

P14.31 Between hospital variation in timings to multidisciplinary glioblastoma care in the Dutch Brain Tumor Registry

BACKGROUND Delay in cancer care may adversely affect emotional distress, treatment outcome and survival. Optimal timings in multidisciplinary glioblastoma care are a matter of debate and clear national guidelines only exist for time to neurosurgery. We evaluated the between-hospital variation in timings to neurosurgery and adjuvant radiotherapy and chemotherapy in newly diagnosed glioblastoma patients in the Netherlands. MATERIAL AND METHODS Data were obtained from the nation-wide Dutch Brain Tumor Registry between 2014 and 2018. All adult patients with glioblastoma were included, covering all 18 neurosurgical hospitals, 28 radiotherapy hospitals, and 33 oncology hospitals. Long time-to-surgery (TTS) was defined as >3 weeks from the date of first brain tumor diagnosis to surgery, long time-to-radiotherapy (TTR) as either >4 or >6 weeks after surgery, and long time-to-chemotherapy (TTC) as either >4 or >6 weeks after completion of radiotherapy. Between-hospital variation in standardized rate of long timings was analyzed in funnel plots after case-mix correction. RESULTS A total of 4203 patients were included. Median TTS was 20 days and 52.4% of patients underwent surgery within 3 weeks. Median TTR was 20 days and 24.6% of patients started radiotherapy within 4 weeks and 84.2% within 6 weeks after surgery. Median TTC was 28 days and 62.6% of patients received chemotherapy within 4 weeks and 91.8% within 6 weeks after radiotherapy. After case-mix correction, three (16.7%) neurosurgical hospitals had significantly more patients with longer than expected TTS. Three (10.7%) and one (3.6%) radiotherapy hospitals had significantly more patients with longer than expected TTR for >4 and >6 weeks, respectively. In seven (21.2%) chemotherapy hospitals, significantly less patients with TTC >4 weeks were observed than expected. In four (12.1%) chemotherapy hospitals, significantly more patients with TTC >4 weeks were observed than expected. CONCLUSION Between-hospital variation in timings to multidisciplinary treatment was observed in glioblastoma care in the Netherlands. A substantial percentage of patients experienced timings longer than anticipated.