Decision support systems help physicians and also play an important role in medical decision-making. They are based on different models, and the best of them are providing an explanation together with an accurate, reliable and quick response. This paper presents a decision support tool for the detection of breast cancer based on three types of decision tree classifiers. They are single decision tree (SDT), boosted decision tree (BDT) and decision tree forest (DTF). Decision tree classification provides a rapid and effective method of categorizing data sets. Decision-making is performed in two stages: training the classifiers with features from Wisconsin breast cancer data set, and then testing. The performance of the proposed structure is evaluated in terms of accuracy, sensitivity, specificity, confusion matrix and receiver operating characteristic (ROC) curves. The results showed that the overall accuracies of SDT and BDT in the training phase achieved 97.07 % with 429 correct classifications and 98.83 % with 437 correct classifications, respectively. BDT performed better than SDT for all performance indices than SDT. Value of ROC and Matthews correlation coefficient (MCC) for BDT in the training phase achieved 0.99971 and 0.9746, respectively, which was superior to SDT classifier. During validation phase, DTF achieved 97.51 %, which was superior to SDT (95.75 %) and BDT (97.07 %) classifiers. Value of ROC and MCC for DTF achieved 0.99382 and 0.9462, respectively. BDT showed the best performance in terms of sensitivity, and SDT was the best only considering speed.

Decision tree classifiers for automated medical diagnosis

Among cancers, breast cancer causes second most number of deaths in women. To reduce the high number of unnecessary breast biopsies, several computer-aided diagnosis systems have been proposed in the last years. These systems help physicians in their decision to perform a breast biopsy on a suspicious lesion seen in a mammogram or to perform a short-term follow-up examination instead. In clinical diagnosis, the use of artificial intelligent techniques as neural networks has shown great potential in this field. In this paper, three classification algorithms, multi-layer perceptron (MLP), radial basis function (RBF) and probabilistic neural networks (PNN), are applied for the purpose of detection and classification of breast cancer. Decision making is performed in two stages: training the classifiers with features from Wisconsin Breast Cancer database and then testing. The performance of the proposed structure is evaluated in terms of sensitivity, specificity, accuracy and ROC. The results revealed that PNN was the best classifiers by achieving accuracy rates of 100 and 97.66 % in both training and testing phases, respectively. MLP was ranked as the second classifier and was capable of achieving 97.80 and 96.34 % classification accuracy for training and validation phases, respectively, using scaled conjugate gradient learning algorithm. However, RBF performed better than MLP in the training phase, and it has achieved the lowest accuracy in the validation phase.

Probabilistic neural network for breast cancer classification

Medical imaging techniques have widely been in use in the diagnosis and detection of breast cancer. The drawback of applying these techniques is the large time consumption in the manual diagnosis of each image pattern by a professional radiologist. Automated classifiers could substantially upgrade the diagnosis process, in terms of both accuracy and time requirement by distinguishing benign and malignant patterns automatically. Neural network (NN) plays an important role in this respect, especially in the application of breast cancer detection. Despite the large number of publications that describe the utilization of NN in various medical techniques, only a few reviews are available that guide the development of these algorithms to enhance the detection techniques with respect to specificity and sensitivity. The purpose of this review is to analyze the contents of recently published literature with special attention to techniques and states of the art of NN in medical imaging. We discuss the usage of NN in four different medical imaging applications to show that NN is not restricted to few areas of medicine. Types of NN used, along with the various types of feeding data, have been reviewed. We also address hybrid NN adaptation in breast cancer detection.

/pdf/artificial-neural-networks-in-image-processing-for-early-3pisz44tvr.pdf

Artificial Neural Networks in Image Processing for Early Detection of Breast Cancer

Mammography is the most widely used tool for the early detection of breast cancer. Computer-based algorithms can be developed to improve diagnostic information in mammograms and assist the radiologist to improve diagnostic accuracy. In this paper, we propose a novel computer aided technique to classify abnormalities in mammograms using fusion of local and global features. The objective of this work is to test the effectiveness of combined use of local and global features in detecting abnormalities in mammograms. Local features used in the system are Chebyshev moments and Haralick’s gray level co-occurrence matrix based texture features. Global features used are Laws texture energy measures, Gabor based texture energy measures and fractal dimension. All types of abnormalities namely clusters of microcalcifications, circumscribed masses, spiculated masses, architectural distortions and ill-defined masses are considered. A support vector machine classifier is designed to classify the samples into abnormal and normal classes. It is observed that combined use of local and global features has improved classification accuracy from 88.75% to 93.17%.

/pdf/fusion-of-local-and-global-features-for-classification-of-19z34bo71f.pdf

Fusion of local and global features for classification of abnormality in mammograms

This paper presents a method for 3D image reconstruction, which is one of the most attractive avenues in digital image processing techniques, especially due to its application in biomedical imaging. The diversity and complexity of tumor cells makes it very challenging to visualize tumor present in magnetic resonance image (MRI) data. This paper involves implementing various steps of extracting the tumor from the 2D slices of MRI brain images by OTSUs threshold technique and various morphological operations and designing software for reconstructing 3D image from a set of 2D tumor images. Extensive use of custom made user interface that provides for ease of user interaction and visualization of reconstructed data. The volume of the tumor is also estimated based on the computation of these images. Doctors and Radiologists can now prepare and image thousands of samples and save on time per day using this automation.

A systematic algorithm for 3-D reconstruction of MRI based brain tumors using morphological operators and bicubic interpolation

Microcalcifications in mammogram have been mainly targeted as a reliable earliest sign of breast cancer and their early detection is vital to improve its prognosis. Since their size is very small and may be easily overlooked by the examining radiologist, computer-based detection output can assist the radiologist to improve the diagnostic accuracy. In this paper, we have proposed an algorithm for detecting microcalcification in mammogram. The proposed microcalcification detection algorithm involves mammogram quality enhancement using multirresolution analysis based on the dyadic wavelet transform and microcalcification detection by fuzzy shell clustering. It may be possible to detect nodular components such as microcalcification accurately by introducing shape information. The effectiveness of the proposed algorithm for microcalcification detection is confirmed by experimental results.

Detection of Microcalcification in Mammograms Using Wavelet Transform and Fuzzy Shell Clustering

https://cyberleninka.org/article/n/331158.pdf

Microcalcification detection in digital mammograms using novel filter bank

In this work, we aimed to study the storm waves generated by Hurricane Ida (2021) which made landfall along the Louisiana coast of the Gulf of Mexico basin. Hurricane Ida was classified as a category 4 (Saffir-Simpson hurricane scale) hurricane by the National Hurricane Center (NHC) and its maximum wind speed reached a peak of 130 kt (67m/s). To study the impact of hurricane winds on waves, two different atmospheric forcing products, ECMWF’s ERA5 and NOAA’s HRRR wind data sets were used. The third-generation wave model WAVEWATCH III (WW3) was used to model the storm waves on a nested grid approach with three different grid resolutions. The simulations were carried out for the entire month of August 2021 and the estimated wave parameters were then compared with observed data obtained at several stations. It was intuitively observed that the results obtained through the atmospheric forcing of ERA5 and HRRR closely correlated with the observed values for non-hurricane conditions. However, during the peak wind conditions, wind fields were better represented in the HRRR model, and it also significantly improved the wave estimates. Through further spatial analyses of wind and wave fields, it was also identified that ERA5 is successful in representing the normal atmospheric conditions and is recommended to be used for studies on long-term seasonal analyses. However, during a hurricane scenario, ERA5 underestimates the wind fields and therefore, HRRR atmospheric forcing is recommended for extreme wave forecasting/modeling in coastal waters.

Storm Wave Modeling in the Gulf of Mexico using nested WAVEWATCH III Model: A Case Study for Hurricane Ida (2021)

Forecast of strong winds and high waves during extreme tropical storms is an essential prerequisite for all marine activities over the Northern Gulf of Mexico (NGoM); including safe navigation, construction and maintenance of coastal infrastructure, understanding of coastal and estuarine morphodynamics for coastal zone management, operation of fishery related and shellfish industries and marine eco-system preservation. This study aims to evaluate sensitivity of wave dynamics in wave model over the NGoM as affected by different wind forcings, using Weather Research Forecasting (WRF) meso-scale model and other wind products to improve the wind representation and consequently, the accuracy of estimated wave characteristics. To this end, the Simulating WAve Nearshore (SWAN) wave model is adopted to simulate wave generation and transformation over the entire GoM, with a focus on NGoM. A detailed SWAN sensitivity analysis is carried out to determine optimal model configuration. Primary model verification was considered against selected data to evaluate the sufficiency of available input wind data from different sources, as well as wave model performance. Necessary improvements were implemented to the input wind data conducting wind modeling to improve the wave model results. Analysis of the statistical parameters shows the improvements due to the adopted wind modeling approach for accurate wave modeling in the study area during tropical storm events.

C. Gowri Shankar

Papers

Detection of Microcalcification in Mammograms Using Wavelet Transform and Fuzzy Shell Clustering

Microcalcification detection in digital mammograms using novel filter bank

Storm Wave Modeling in the Gulf of Mexico using nested WAVEWATCH III Model: A Case Study for Hurricane Ida (2021)

Improvement of Wind Data Quality for Extreme Wave Generation over the Northern Gulf of Mexico