Alzheimer’s Disease (AD) is the most common cause of dementia globally. It steadily worsens from mild to severe, impairing one’s ability to complete any work without assistance. It begins to outstrip due to the population ages and diagnosis timeline. For classifying cases, existing approaches incorporate medical history, neuropsychological testing, and Magnetic Resonance Imaging (MRI), but efficient procedures remain inconsistent due to lack of sensitivity and precision. The Convolutional Neural Network (CNN) is utilized to create a framework that can be used to detect specific Alzheimer’s disease characteristics from MRI images. By considering four stages of dementia and conducting a particular diagnosis, the proposed model generates high-resolution disease probability maps from the local brain structure to a multilayer perceptron and provides accurate, intuitive visualizations of individual Alzheimer’s disease risk. To avoid the problem of class imbalance, the samples should be evenly distributed among the classes. The obtained MRI image dataset from Kaggle has a major class imbalance problem. A DEMentia NETwork (DEMNET) is proposed to detect the dementia stages from MRI. The DEMNET achieves an accuracy of 95.23%, Area Under Curve (AUC) of 97% and Cohen’s Kappa value of 0.93 from the Kaggle dataset, which is superior to existing methods. We also used the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset to predict AD classes in order to assess the efficacy of the proposed model.

/pdf/demnet-a-deep-learning-model-for-early-diagnosis-of-46kt1xnwz3.pdf

DEMNET: A Deep Learning Model for Early Diagnosis of Alzheimer Diseases and Dementia From MR Images

The high mortality rate associated with brain tumors requires early detection in the early stages to treat and reduce mortality. Due to the complexity of brain tissue, manual diagnosis of the brain and tumor tissues is very time‐consuming and operator dependent. Furthermore, there is a need for experts who can review the images to detect these effects, rendering traditional methods inefficient in their presence. Therefore, the use of automated procedures for the careful examination of tumors can prove useful. In this study, a new metaheuristic‐based system is presented for the early detection of brain tumors. The proposed method implements three main steps, namely tumor segmentation, feature extraction, and classification based on a deep belief network. An improved version of the seagull optimization algorithm is adopted for optimal selection of the features and classification of the images. The simulation results of the proposed method are compared with a few existing methods. The final results demonstrate that the proposed method exhibits superior performance in terms of the CDR, FAR, and FRR indices compared with the other methods.

Brain tumor diagnosis based on metaheuristics and deep learning

New brain tumor classification method based on an improved version of whale optimization algorithm

An apparatus comprises a receiver (203) which receives a microphone signal from a microphone (201) where the microphone signal comprises a test signal component corresponding to an audio test signal. A divider (215) divides the microphone signal into a plurality of test interval signal components, each of which corresponds to the microphone signal in a time interval. A set processor (217) generates sets of test interval signal components and a similarity processor (219) generates a similarity value for each set. An interference estimator (221) determines an interference measure for individual test interval signal components in response to the similarity values. The interference measure may be used to select signal segments that can be used to adapt an audio processing algorithm which is applied to the microphone signal, such as e.g. speech enhancement or echo cancellation. The approach may allow for a reliable interference estimate to be generated while maintaining low complexity.

Method and apparatus for audio interference estimation

Audio classification serves as the fundamental step towards the rapid growth in audio data volume. Due to the increasing size of the multimedia sources speec h and music classification is one of the most impor tant issues for multimedia information retrieval. In thi s work a speech/music discrimination system is deve loped which utilizes the Discrete Wavelet Transform (DWT) as the acoustic feature. Multi resolution analysis is the most significant statistical way to extract the features from the input signal and in this study, a method is deployed to model the extracted wavelet feature. Su pport Vector Machines (SVM) are based on the principle of structural risk minimization. SVM is a pplied to classify audio into their classes namely speech and music, by learning from training data. Then the proposed method extends the application of Gaussia n Mixture Models (GMM) to estimate the probability density function using maximum likelihood decision methods. The system shows significant results with an accuracy of 94.5%.

/pdf/speech-music-classification-using-wavelet-based-feature-3u1qztjhx5.pdf

Speech/music classification using wavelet based feature extraction techniques

Diagnosing the brain tumor from Magnetic Resonance Imaging (MRI) in Computer-Aided Diagnosis (CAD) applications is one of the challenging task in medical image processing. Traditionally many segmentation methods are used to address this issue. This paper introduces a segmentation method along with image fusion. Here a Discrete Wavelet Transform (DWT) method is chosen, for image fusion followed by segmentation using Support Vector Machine (SVM) for detecting the abnormality region. The types of MRI images considered here include T1-weighted (T1-w), T2-weighted (T2-w) and FLAIR images. The various fusion combinations are T1-w and T2-w, T1-w and FLAIR, T2-w and FLAIR. Experimental results suggest that on an average, fusion-based segmented result is superior to non-fusion-based segmented result.

Fusion-Based Segmentation Technique for Improving the Diagnosis of MRI Brain Tumor in CAD Applications

Problem statement: The received voice signal in mobile communication is often disturbed
by background noise and hence there is a need for good noise reduction methods for enhancing
Speech. It is well known that denoising is a compromise between the removal of the largest possible
amount of noise and the preservation of signal integrity. To address this issue, a new method for
enhancing speech from background interference is presented in this study by fusing dual band spectral
subtraction with adaptive noise estimator and wavelet packet based thresholding method. Approach:
The proposed system uses the combination of dual band Spectral Subtraction method with adaptive
noise estimator for pre-processing, in order to initially reduce the noise level and further the quality of
speech is improved by Wavelet Packet Transform (WPT) based level dependent thresholding method.
The threshold value is determined by using Stein’s Unbiased Risk Estimator (SURE) and hard, soft,
Garrotte, µ-law and a proposed modified soft thresholding functions are considered for denoising.
Results: The proposed method was investigated by ten different clean speech samples (five male and
five female) taken from TIMIT database and thirteen different noise sources to degrade the speech
artificially and the energy level of the noise is scaled such that the overall SNR of the noisy speech is
maintained at -5, 0,5,10 and 15 dB and finally the results are evaluated using objective and subjective
measures. Conclusion/Recommendations: It is suggested from the experimental results that the
proposed scheme gives improved spectral performance, reflects in better speech quality in all types of
noisy environment. For better speech enhancement in noise dominated regions, the system efficiency is
further improved by fusing threshold values for wavelet denoising.

A New Robust Hybrid Approach to Enhance Speech in Mobile Communication Systems

This paper describes the concert study of different segmentation techniques for magnetic resonance imaging (MRI) and positron emission tomography (PET) brain tumor detection. Many segmentation methodologies are available for diagnosis of medical images. In general, the segmentation approaches are classified based on the properties of image such as to detect the discontinuities and to detect the similarities. In medical field, it is essential to distinguish the abnormal region at an early stage, which is a difficult task for neuroradiologist. Hence to solve this problem, performance of several segmentation techniques is analyzed. The segmentation methods considered here are gray-level co-occurrence matrix (GLCM), gradient vector flow (GVF), wavelet transform (WT), ant colony optimization (ACO) and artificial bee colony (ABC). The various performance metrics are used to evaluate the considered segmentation techniques. The attained results suggest that in case of MRI images, GVF method is acting out better, and with respect to PET images, the concert of both ABC and ACO goes parallel.

Bharathi Deepa

Papers

Fusion-Based Segmentation Technique for Improving the Diagnosis of MRI Brain Tumor in CAD Applications

A New Robust Hybrid Approach to Enhance Speech in Mobile Communication Systems

Recital Study of Different Segmentation Techniques for Brain Tumor Detection