Classifier Model Based on Machine Learning Algorithms: Application to Differential Diagnosis of Suspicious Thyroid Nodules via Sonography

2016 ◽  
Vol 207 (4) ◽  
pp. 859-864 ◽  
Author(s):  
Hongxun Wu ◽  
Zhaohong Deng ◽  
Bingjie Zhang ◽  
Qianyun Liu ◽  
Junyong Chen
Keyword(s):  
Machine Learning ◽  
Differential Diagnosis ◽  
Thyroid Nodules ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Model Based

scholarly journals BI-RADS Reading of Non-Mass Lesions on DCE-MRI and Differential Diagnosis Performed by Radiomics and Deep Learning

2021 ◽  
Vol 11 ◽  
Author(s):  
Jiejie Zhou ◽  
Yan-Lin Liu ◽  
Yang Zhang ◽  
Jeon-Hor Chen ◽  
Freddie J. Combs ◽  
...  
Keyword(s):  
Machine Learning ◽  
Differential Diagnosis ◽  
Deep Learning ◽  
Learning Algorithms ◽  
Regional Distribution ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Mass Lesions ◽  
Morphological Distribution ◽  
Parametric Maps

BackgroundA wide variety of benign and malignant processes can manifest as non-mass enhancement (NME) in breast MRI. Compared to mass lesions, there are no distinct features that can be used for differential diagnosis. The purpose is to use the BI-RADS descriptors and models developed using radiomics and deep learning to distinguish benign from malignant NME lesions.Materials and MethodsA total of 150 patients with 104 malignant and 46 benign NME were analyzed. Three radiologists performed reading for morphological distribution and internal enhancement using the 5th BI-RADS lexicon. For each case, the 3D tumor mask was generated using Fuzzy-C-Means segmentation. Three DCE parametric maps related to wash-in, maximum, and wash-out were generated, and PyRadiomics was applied to extract features. The radiomics model was built using five machine learning algorithms. ResNet50 was implemented using three parametric maps as input. Approximately 70% of earlier cases were used for training, and 30% of later cases were held out for testing.ResultsThe diagnostic BI-RADS in the original MRI report showed that 104/104 malignant and 36/46 benign lesions had a BI-RADS score of 4A–5. For category reading, the kappa coefficient was 0.83 for morphological distribution (excellent) and 0.52 for internal enhancement (moderate). Segmental and Regional distribution were the most prominent for the malignant group, and focal distribution for the benign group. Eight radiomics features were selected by support vector machine (SVM). Among the five machine learning algorithms, SVM yielded the highest accuracy of 80.4% in training and 77.5% in testing datasets. ResNet50 had a better diagnostic performance, 91.5% in training and 83.3% in testing datasets.ConclusionDiagnosis of NME was challenging, and the BI-RADS scores and descriptors showed a substantial overlap. Radiomics and deep learning may provide a useful CAD tool to aid in diagnosis.

scholarly journals Deep-ACTINet: End-to-End Deep Learning Architecture for Automatic Sleep-Wake Detection Using Wrist Actigraphy

Electronics ◽  
2019 ◽  
Vol 8 (12) ◽  
pp. 1461 ◽  
Author(s):  
Taeheum Cho ◽  
Unang Sunarya ◽  
Minsoo Yeo ◽  
Bosun Hwang ◽  
Yong Seo Koo ◽  
...  
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Sleep Disorders ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Feature Engineering ◽  
Model Based ◽  
Wrist Actigraphy ◽  
End To End ◽  
Wake Detection

Sleep scoring is the first step for diagnosing sleep disorders. A variety of chronic diseases related to sleep disorders could be identified using sleep-state estimation. This paper presents an end-to-end deep learning architecture using wrist actigraphy, called Deep-ACTINet, for automatic sleep-wake detection using only noise canceled raw activity signals recorded during sleep and without a feature engineering method. As a benchmark test, the proposed Deep-ACTINet is compared with two conventional fixed model based sleep-wake scoring algorithms and four feature engineering based machine learning algorithms. The datasets were recorded from 10 subjects using three-axis accelerometer wristband sensors for eight hours in bed. The sleep recordings were analyzed using Deep-ACTINet and conventional approaches, and the suggested end-to-end deep learning model gained the highest accuracy of 89.65%, recall of 92.99%, and precision of 92.09% on average. These values were approximately 4.74% and 4.05% higher than those for the traditional model based and feature based machine learning algorithms, respectively. In addition, the neuron outputs of Deep-ACTINet contained the most significant information for separating the asleep and awake states, which was demonstrated by their high correlations with conventional significant features. Deep-ACTINet was designed to be a general model and thus has the potential to replace current actigraphy algorithms equipped in wristband wearable devices.

Comparison between linear and nonlinear machine-learning algorithms for the classification of thyroid nodules

2019 ◽  
Vol 113 ◽  
pp. 251-257 ◽  
Author(s):  
Fu-sheng Ouyang ◽  
Bao-liang Guo ◽  
Li-zhu Ouyang ◽  
Zi-wei Liu ◽  
Shao-jia Lin ◽  
...  
Keyword(s):  
Machine Learning ◽  
Thyroid Nodules ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Linear And Nonlinear

scholarly journals Density Based Spatial Clustering Application with Noise by Varying Densities

2019 ◽  
Vol 8 (4) ◽  
pp. 5886-5891
Keyword(s):  
Machine Learning ◽  
Spatial Clustering ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Cluster Algorithms ◽  
Model Based ◽  
Dbscan Algorithm ◽  
Density Values ◽  
Original Size

Cluster algorithms are used for grouping up of similar points to form a cluster. It has seen mostly in Machine Learning algorithms. The most popular density-based algorithm is DBSCAN. DBSCAN can find the clusters, irrespective of its shapes and sizes of a cluster. DBSCAN algorithm can easily detect the noise in a clustering dataset. In the proposed algorithm we developed a model based on the existing dbscan algorithm. In the developed algorithm we focus mainly on the epsilon parameter value. Whenever the dbscan algorithm fails to form a cluster we increase the epsilon value by half of its original size. We repeat this step until a cluster is formed. Whenever a cluster is newly formed we change existing epsilon parameter value by adding the 10 percent of the previous used epsilon parameter value. We use epsilon for varying the density of a cluster. So, we can use the dbscan algorithm with the varying density values for developing a cluster. We applied this algorithm on the various datasets.

Sign in / Sign up

Export Citation Format

Share Document