Optimization of breast mass classification using sequential forward floating selection (SFFS) and a support vector machine (SVM) model

The top 100 most-cited articles on artificial intelligence in breast radiology: a bibliometric analysis

INTRODUCTION

Artificial intelligence (AI) in radiology is a rapidly evolving field. In breast imaging, AI has already been applied in a real-world setting and multiple studies have been conducted in the area. The aim of this analysis is to identify the most influential publications on the topic of artificial intelligence in breast imaging.

METHODS

A retrospective bibliometric analysis was conducted on artificial intelligence in breast radiology using the Web of Science database. The search strategy involved searching for the keywords 'breast radiology' or 'breast imaging' and the various keywords associated with AI such as 'deep learning', 'machine learning,' and 'neural networks'.

RESULTS

From the top 100 list, the number of citations per article ranged from 30 to 346 (average 85). The highest cited article titled 'Artificial Neural Networks In Mammography-Application To Decision-Making In The Diagnosis Of Breast-Cancer' was published in Radiology in 1993. Eighty-three of the articles were published in the last 10 years. The journal with the greatest number of articles was Radiology (n = 22). The most common country of origin was the United States (n = 51). Commonly occurring topics published were the use of deep learning models for breast cancer detection in mammography or ultrasound, radiomics in breast cancer, and the use of AI for breast cancer risk prediction.

CONCLUSION

This study provides a comprehensive analysis of the top 100 most-cited papers on the subject of artificial intelligence in breast radiology and discusses the current most influential papers in the field.

CLINICAL RELEVANCE STATEMENT

This article provides a concise summary of the top 100 most-cited articles in the field of artificial intelligence in breast radiology. It discusses the most impactful articles and explores the recent trends and topics of research in the field.

KEY POINTS

Multiple studies have been conducted on AI in breast radiology. The most-cited article was published in the journal Radiology in 1993. This study highlights influential articles and topics on AI in breast radiology.

Unraveling the interplay of ferroptosis and immune dysregulation in diabetic kidney disease: a comprehensive molecular analysis

BACKGROUND

Diabetic kidney disease (DKD) is a primary microvascular complication of diabetes with limited therapeutic effects. Delving into the pathogenic mechanisms of DKD and identifying new therapeutic targets is crucial. Emerging studies reveal the implication of ferroptosis and immune dysregulation in the pathogenesis of DKD, however, the precise relationship between them remains not fully elucidated. Investigating their interplay is pivotal to unraveling the pathogenesis of diabetic kidney disease, offering insights crucial for targeted interventions and improved patient outcomes.

METHODS

Integrated analysis, Consensus clustering, Machine learning including Generalized Linear Models (GLM), RandomForest (RF), Support Vector Machine (SVM) and Extreme Gradient Boosting (xGB), Artificial neural network (ANN) methods of DKD glomerular mRNA sequencing were performed to screen DKD-related ferroptosis genes.CIBERSORT, ESTIMATE and ssGSEA algorithm were used to assess the infiltration of immune cells between DKD and control groups and in two distinct ferroptosis phenotypes. The ferroptosis hub genes were verified in patients with DKD and in the db/db spontaneous type 2 diabetes mouse model via immunohistochemical and Western blotting analyses in mouse podocyte MPC5 and mesangial SV40-MES-13 cells under high-glucose (HG) conditions.

RESULTS

We obtained 16 differentially expressed ferroptosis related genes and patients with DKD were clustered into two subgroups by consensus clustering. Five ferroptosis genes (DUSP1,ZFP36,PDK4,CD44 and RGS4) were identified to construct a diagnostic model with a good diagnosis performance in external validation. Analysis of immune infiltration revealed immune heterogeneity between DKD patients and controls.Moreover, a notable differentiation in immune landscape, comprised of Immune cells, ESTIMATE Score, Immune Score and Stromal Score was observed between two FRG clusters. GSVA analysis indicated that autophagy, apoptosis and complement activation can participate in the regulation of ferroptosis phenotypes. Experiment results showed that ZFP36 was significantly overexpressed in both tissue and cells while CD44 was on the contrary.Meanwhile,spearman analysis showed both ZFP36 and CD44 has a strong correlation with different immune cells,especially macrophage.

CONCLUSION

The regulation of the immune landscape in DKD is significantly influenced by the focal point on ferroptosis. Newly identified ferroptosis markers, CD44 and ZFP36, are poised to play essential roles through their interactions with macrophages, adding substantial value to this regulatory landscape.

Investigating the impact of Wnt pathway-related genes on biomarker and diagnostic model development for osteoporosis in postmenopausal females

The Wnt signaling pathway is essential for bone development and maintaining skeletal homeostasis, making it particularly relevant in osteoporosis patients. Our study aimed to identify distinct molecular clusters associated with the Wnt pathway and develop a diagnostic model for osteoporosis in postmenopausal Caucasian women. We downloaded three datasets (GSE56814, GSE56815 and GSE2208) related to osteoporosis from the GEO database. Our analysis identified a total of 371 differentially expressed genes (DEGs) between low and high bone mineral density (BMD) groups, with 12 genes associated with the Wnt signaling pathway, referred to as osteoporosis-associated Wnt pathway-related genes. Employing four independent machine learning models, we established a diagnostic model using the 12 osteoporosis-associated Wnt pathway-related genes in the training set. The XGB model showed the most promising discriminative potential. We further validate the predictive capability of our diagnostic model by applying it to three external datasets specifically related to osteoporosis. Subsequently, we constructed a diagnostic nomogram based on the five crucial genes identified from the XGB model. In addition, through the utilization of DGIdb, we identified a total of 30 molecular compounds or medications that exhibit potential as promising therapeutic targets for osteoporosis. In summary, our comprehensive analysis provides valuable insights into the relationship between the osteoporosis and Wnt signaling pathway.

Identification and validation of cuproptosis-related molecular clusters in non-alcoholic fatty liver disease

Non-alcoholic fatty liver disease (NAFLD) is a major chronic liver disease worldwide. Cuproptosis has recently been reported as a form of cell death that appears to drive the progression of a variety of diseases. This study aimed to explore cuproptosis-related molecular clusters and construct a prediction model. The gene expression profiles were obtained from the Gene Expression Omnibus (GEO) database. The associations between molecular clusters of cuproptosis-related genes and immune cell infiltration were investigated using 50 NAFLD samples. Furthermore, cluster-specific differentially expressed genes were identified by the WGCNA algorithm. External datasets were used to verify and screen feature genes, and nomograms, calibration curves and decision curve analysis (DCA) were performed to verify the performance of the prediction model. Finally, a NAFLD-diet mouse model was constructed to further verify the predictive analysis, thus providing new insights into the prediction of NAFLD clusters and risks. The role of cuproptosis in the development of non-alcoholic fatty liver disease and immune cell infiltration was explored. Non-alcoholic fatty liver disease was divided into two cuproptosis-related molecular clusters by unsupervised clustering. Three characteristic genes (ENO3, SLC16A1 and LEPR) were selected by machine learning and external data set validation. In addition, the accuracy of the nomogram, calibration curve and decision curve analysis in predicting NAFLD clusters was also verified. Further animal and cell experiments confirmed the difference in their expression in the NAFLD mouse model and Mouse hepatocyte cell line. The present study explored the relationship between non-alcoholic fatty liver disease and cuproptosis, providing new ideas and targets for individual treatment of the disease.

A multi-stage fusion framework to classify breast lesions using deep learning and radiomics features computed from four-view mammograms

BACKGROUND

Developing computer aided diagnosis (CAD) schemes of mammograms to classify between malignant and benign breast lesions has attracted a lot of research attention over the last several decades. However, unlike radiologists who make diagnostic decisions based on the fusion of image features extracted from multi-view mammograms, most CAD schemes are single-view-based schemes, which limit CAD performance and clinical utility.

PURPOSE

This study aims to develop and test a novel CAD framework that optimally fuses information extracted from ipsilateral views of bilateral mammograms using both deep transfer learning (DTL) and radiomics feature extraction methods.

METHODS

An image dataset containing 353 benign and 611 malignant cases is assembled. Each case contains four images: the craniocaudal (CC) and mediolateral oblique (MLO) view of the left and right breast. First, we extract four matching regions of interest (ROIs) from images that surround centers of two suspicious lesion regions seen in CC and MLO views, as well as matching ROIs in the contralateral breasts. Next, the handcrafted radiomics (HCRs) features and VGG16 model-generated automated features are extracted from each ROI resulting in eight feature vectors. Then, after reducing feature dimensionality and quantifying the bilateral and ipsilateral asymmetry of four ROIs to yield four new feature vectors, we test four fusion methods to build three support vector machine (SVM) classifiers by an optimal fusion of asymmetrical image features extracted from four view images.

RESULTS

Using a 10-fold cross-validation method, results show that a SVM classifier trained using an optimal fusion of four view images yields the highest classification performance (AUC = 0.876 ± 0.031), which significantly outperforms SVM classifiers trained using one projection view alone, AUC = 0.817 ± 0.026 and 0.792 ± 0.026 for the CC and MLO view of bilateral mammograms, respectively (p < 0.001).

CONCLUSIONS

The study demonstrates that the shift from single-view CAD to four-view CAD and the inclusion of both DTL and radiomics features significantly increases CAD performance in distinguishing between malignant and benign breast lesions.

Single cell and bulk transcriptome analysis identified oxidative stress response-related features of Hepatocellular Carcinoma

Background: Hepatocellular Carcinoma (HCC) is a common lethal digestive system tumor. The oxidative stress mechanism is crucial in the HCC genesis and progression. Methods: Our study analyzed single-cell and bulk sequencing data to compare the microenvironment of non-tumor liver tissues and HCC tissues. Through these analyses, we aimed to investigate the effect of oxidative stress on cells in the HCC microenvironment and identify critical oxidative stress response-related genes that impact the survival of HCC patients. Results: Our results showed increased oxidative stress in HCC tissue compared to non-tumor tissue. Immune cells in the HCC microenvironment exhibited higher oxidative detoxification capacity, and oxidative stress-induced cell death of dendritic cells was attenuated. HCC cells demonstrated enhanced communication with immune cells through the MIF pathway in a highly oxidative hepatoma microenvironment. Meanwhile, using machine learning and Cox regression screening, we identified PRDX1 as a predictor of early occurrence and prognosis in patients with HCC. The expression level of PRDX1 in HCC was related to dysregulated ribosome biogenesis and positively correlated with the expression of immunological checkpoints (PDCD1LG2, CTLA4, TIGIT, LAIR1). High PRDX1 expression in HCC patients correlated with better sensitivity to immunotherapy agents such as sorafenib, IGF-1R inhibitor, and JAK inhibitor. Conclusion: In conclusion, our study unveiled variations in oxidative stress levels between non-tumor liver and HCC tissues. And we identified oxidative stress gene markers associated with hepatocarcinogenesis development, offering novel insights into the oxidative stress response mechanism in HCC.

Development of survival predictors for high-grade serous ovarian cancer based on stable radiomic features from computed tomography images

Less than 35% of advanced patients with high-grade serous ovarian cancer (HGSOC) survive for 5 years after diagnosis. Here, we developed radiomics-based models to predict HGSOC clinical outcomes using preoperative contrast-enhanced computed tomography (CECT) images. 891 radiomics features were extracted between primary, metastatic, or lymphatic lesions from preoperative venous phase CECT images of 217 patients with HGSOC. A heuristic method, Frequency Appearance in Multiple Univariate preScreening (FAMUS), was proposed to identify stable and task-relevant radiomic features. Using FAMUS, we constructed predictive models of overall survival and disease-free survival in patients with HGSOC based on these stable radiomic features. According to their CT images, patients with HGSOC can be accurately stratified into high-risk or low-risk groups for cancer-related death within 2-6 years or for likely recurrence within 1-5 years. These radiomic models provide convincing and reliable non-invasive markers for individualized prognostic evaluation and clinical decision-making for patients with HGSOC.

•

Frequency Appearance in Multiple Univariate preScreening (FAMUS) identifies stable and task-relevant radiomic features from computed tomography (CT) images

•

Radiomics-based signatures are highly predictive of the clinical outcome of high-grade serous ovarian cancer (HGSOC)

•

FAMUS improves the prognostic performance of radiomics-based prediction models

•

Developed radiomic models can help clinicians tailor treatment plans for HGSOC

Comparison of Machine Learning Methods Using Spectralis OCT for Diagnosis and Disability Progression Prognosis in Multiple Sclerosis

Machine learning approaches in diagnosis and prognosis of multiple sclerosis (MS) were analysed using retinal nerve fiber layer (RNFL) thickness, measured by optical coherence tomography (OCT). A cross-sectional study (72 MS patients and 30 healthy controls) was used for diagnosis. These 72 MS patients were involved in a 10-year longitudinal follow-up study for prognostic purposes. Structural measurements of RNFL thickness were performed using different Spectralis OCT protocols: fast macular thickness protocol to measure macular RNFL, and fast RNFL thickness protocol and fast RNFL-N thickness protocol to measure peripapillary RNFL. Binary classifiers such as multiple linear regression (MLR), support vector machines (SVM), decision tree (DT), k-nearest neighbours (k-NN), Naïve Bayes (NB), ensemble classifier (EC) and long short-term memory (LSTM) recurrent neural network were tested. For MS diagnosis, the best acquisition protocol was fast macular thickness protocol using k-NN (accuracy: 95.8%; sensitivity: 94.4%; specificity: 97.2%; precision: 97.1%; AUC: 0.958). For MS prognosis, our model with a 3-year follow up to predict disability progression 8 years later was the best predictive model. DT performed best for fast macular thickness protocol (accuracy: 91.3%; sensitivity: 90.0%; specificity: 92.5%; precision: 92.3%; AUC: 0.913) and SVM for fast RNFL-N thickness protocol (accuracy: 91.3%; sensitivity: 87.5%; specificity: 95.0%; precision: 94.6%; AUC: 0.913). This work concludes that measurements of RNFL thickness obtained with Spectralis OCT have a good ability to diagnose MS and to predict disability progression in MS patients. This machine learning approach would help clinicians to have valuable information.

Improving mammography lesion classification by optimal fusion of handcrafted and deep transfer learning features

Objective.Handcrafted radiomics features or deep learning model-generated automated features are commonly used to develop computer-aided diagnosis schemes of medical images. The objective of this study is to test the hypothesis that handcrafted and automated features contain complementary classification information and fusion of these two types of features can improve CAD performance.Approach.We retrospectively assembled a dataset involving 1535 lesions (740 malignant and 795 benign). Regions of interest (ROI) surrounding suspicious lesions are extracted and two types of features are computed from each ROI. The first one includes 40 radiomic features and the second one includes automated features computed from a VGG16 network using a transfer learning method. A single channel ROI image is converted to three channel pseudo-ROI images by stacking the original image, a bilateral filtered image, and a histogram equalized image. Two VGG16 models using pseudo-ROIs and 3 stacked original ROIs without pre-processing are used to extract automated features. Five linear support vector machines (SVM) are built using the optimally selected feature vectors from the handcrafted features, two sets of VGG16 model-generated automated features, and the fusion of handcrafted and each set of automated features, respectively.Main Results.Using a 10-fold cross-validation, the fusion SVM using pseudo-ROIs yields the highest lesion classification performance with area under ROC curve (AUC = 0.756 ± 0.042), which is significantly higher than those yielded by other SVMs trained using handcrafted or automated features only (p < 0.05).Significance.This study demonstrates that both handcrafted and automated futures contain useful information to classify breast lesions. Fusion of these two types of features can further increase CAD performance.

An effective fine grading method of BI-RADS classification in mammography

PURPOSE

Mammography is an important imaging technique for the detection of early breast cancer. Doctors classify mammograms as Breast Imaging Reporting and Data Systems (BI-RADS). This study aims to provide an intelligent BI-RADS grading prediction method, which can help radiologists and clinicians to distinguish the most challenging 4A, 4B, and 4C cases in mammography.

METHODS

Firstly, the breast region, the lesion region, and the corresponding region in the contralateral breast were extracted. Four categories of features were extracted from the original images and the images after the wavelet transform. Secondly, an optimized sequential forward floating selection (SFFS) was used for feature selection. Finally, a two-layer classifier integration was employed for fine grading prediction. 45 cases from the hospital and 500 cases from Digital Database for Screening Mammography (DDSM) database were used for evaluation.

RESULTS

The classification performance of the support vector machine (SVM), Bayes, and random forest is very close on the 45 testing set, with the area under the receiver operating characteristic curve (AUC) of 0.978, 0.967, and 0.968. On the DDSM set, the AUC achieves 0.931, 0.938, and 0.874. Using the mean probability prediction, the AUC on the two datasets reaches 0.998 and 0.916. However, they are all significantly higher than the doctors' diagnosis, with the AUC of 0.807 and 0.725.

CONCLUSIONS

A BI-RADS fine grading (2, 3, 4A, 4B, 4C, 5) prediction model was proposed. Through the evaluation from different datasets, the performance is proved higher than that of the doctors, which may provide great help for clinical BI-RADS classification diagnosis. Therefore, our method can produce more effective and reliable results.

Cross-modality image feature fusion diagnosis in breast cancer

Considering the complementarity of mammography and breast MRI, the research of feature fusion diagnosis based on cross-modality images was explored to improve the accuracy of breast cancer diagnosis. 201 patients with both mammography and breast MRI were collected retrospectively, including 117 cases of benign lesions and 84 cases of malignant ones. Two feature optimization strategies of sequential floating forward selection (SFFS), SFFS-1 and SFFS-2, were defined based on the sequential floating forward selection method. Each strategy was used to analyze the diagnostic performance of single-modality images and then to study the feature fusion diagnosis of cross-modality images. Three feature fusion approaches were compared: optimizing MRI features and then fusing those of mammography; optimizing mammography features and then fusing those of MRI; selecting the effective features from the whole feature set (mammography and MRI). Support vector machine, Naive Bayes, and K-nearest neighbor were employed as the classifiers and were finally integrated to get better performance. The average accuracy and area under the ROC curve (AUC) of MRI (88.56%, 0.9 for SFFS-1, 88.39%, 0.89 for SFFS-2) were better than mammography (84.25%, 0.84 for SFFS-1, 80.43%, 0.80 for SFFS-2). Furthermore, compared with a single modality, the average accuracy and AUC of cross-modality feature fusion can improve from 85.40% and 0.86 to 89.66% and 0.91. Classifier integration improved the accuracy and AUC from 90.49%, 0.92 to 92.37%, and 0.97. Cross-modality image feature fusion can achieve better diagnosis performance than a single modality. Feature selection strategy SFFS-1 has better efficiency than SFFS-2. Classifier integration can further improve diagnostic accuracy.

Machine learning in diagnosis and disability prediction of multiple sclerosis using optical coherence tomography

BACKGROUND

Multiple sclerosis (MS) is a neurodegenerative disease that affects the central nervous system, especially the brain, spinal cord, and optic nerve. Diagnosis of this disease is a very complex process and generally requires a lot of time. In addition, treatments are applied without any information on the disability course in each MS patient. For these two reasons, the objective of this study was to improve the MS diagnosis and predict the long-term course of disability in MS patients based on clinical data and retinal nerve fiber layer (RNFL) thickness, measured by optical coherence tomography (OCT).

MATERIAL AND METHODS

A total of 104 healthy controls and 108 MS patients, 82 of whom had a 10-year follow-up, were enrolled. Classification algorithms such as multiple linear regression (MLR), support vector machines (SVM), decision tree (DT), k-nearest neighbours (k-NN), Naïve Bayes (NB), ensemble classifier (EC) and long short-term memory (LSTM) recurrent neural network were tested to develop two predictive models: MS diagnosis model and MS disability course prediction model.

RESULTS

For MS diagnosis, the best result was obtained using EC (accuracy: 87.7%; sensitivity: 87.0%; specificity: 88.5%; precision: 88.7%; AUC: 0.8775). In line with this good performance, the accuracy was 85.4% using k-NN and 84.4% using SVM. And, for long-term prediction of MS disability course, LSTM recurrent neural network was the most appropriate classifier (accuracy: 81.7%; sensitivity: 81.1%; specificity: 82.2%; precision: 78.9%; AUC: 0.8165). The use of MLR, SVM and k-NN also showed a good performance (AUC ≥ 0.8).

CONCLUSIONS

This study demonstrated that machine learning techniques, using clinical and OCT data, can help establish an early diagnosis and predict the course of MS. This advance could help clinicians select more specific treatments for each MS patient. Therefore, our findings underscore the potential of RNFL thickness as a reliable MS biomarker.

Applying a random projection algorithm to optimize machine learning model for predicting peritoneal metastasis in gastric cancer patients using CT images

BACKGROUND AND OBJECTIVE

Non-invasively predicting the risk of cancer metastasis before surgery can play an essential role in determining which patients can benefit from neoadjuvant chemotherapy. This study aims to investigate and test the advantages of applying a random projection algorithm to develop and optimize a radiomics-based machine learning model to predict peritoneal metastasis in gastric cancer patients using a small and imbalanced computed tomography (CT) image dataset.

METHODS

A retrospective dataset involving CT images acquired from 159 patients is assembled, including 121 and 38 cases with and without peritoneal metastasis, respectively. A computer-aided detection scheme is first applied to segment primary gastric tumor volumes and initially compute 315 image features. Then, five gradients boosting machine (GBM) models embedded with five feature selection methods (including random projection algorithm, principal component analysis, least absolute shrinkage, and selection operator, maximum relevance and minimum redundancy, and recursive feature elimination) along with a synthetic minority oversampling technique, are built to predict the risk of peritoneal metastasis. All GBM models are trained and tested using a leave-one-case-out cross-validation method.

RESULTS

Results show that the GBM model embedded with a random projection algorithm yields a significantly higher prediction accuracy (71.2%) than the other four GBM models (p<0.05). The precision, sensitivity, and specificity of this optimal GBM model are 65.78%, 43.10%, and 87.12%, respectively.

CONCLUSIONS

This study demonstrates that CT images of the primary gastric tumors contain discriminatory information to predict the risk of peritoneal metastasis, and a random projection algorithm is a promising method to generate optimal feature vector, improving the performance of machine learning based prediction models.

Applying a Random Projection Algorithm to Optimize Machine Learning Model for Breast Lesion Classification

OBJECTIVE

Since computer-aided diagnosis (CAD) schemes of medical images usually computes large number of image features, which creates a challenge of how to identify a small and optimal feature vector to build robust machine learning models, the objective of this study is to investigate feasibility of applying a random projection algorithm (RPA) to build an optimal feature vector from the initially CAD-generated large feature pool and improve performance of machine learning model.

METHODS

We assemble a retrospective dataset involving 1,487 cases of mammograms in which 644 cases have confirmed malignant mass lesions and 843 have benign lesions. A CAD scheme is first applied to segment mass regions and initially compute 181 features. Then, support vector machine (SVM) models embedded with several feature dimensionality reduction methods are built to predict likelihood of lesions being malignant. All SVM models are trained and tested using a leave-one-case-out cross-validation method. SVM generates a likelihood score of each segmented mass region depicting on one-view mammogram. By fusion of two scores of the same mass depicting on two-view mammograms, a case-based likelihood score is also evaluated.

RESULTS

Comparing with the principle component analyses, nonnegative matrix factorization, and Chi-squared methods, SVM embedded with RPA yielded a significantly higher case-based lesion classification performance with the area under ROC curve of 0.84 ± 0.01 (p<0.02).

CONCLUSION

The study demonstrates that RPA is a promising method to generate optimal feature vectors and improve SVM performance.

SIGNIFICANCE

This study presents a new method to develop CAD schemes with significantly higher and robust performance.

Developing a new radiomics-based CT image marker to detect lymph node metastasis among cervical cancer patients

BACKGROUND AND OBJECTIVE

In diagnosis of cervical cancer patients, lymph node (LN) metastasis is a highly important indicator for the following treatment management. Although CT/PET (i.e., computed tomography/positron emission tomography) examination is the most effective approach for this detection, it is limited by the high cost and low accessibility, especially for the rural areas in the U.S.A. or other developing countries. To address this challenge, this investigation aims to develop and test a novel radiomics-based CT image marker to detect lymph node metastasis for cervical cancer patients.

METHODS

A total of 1,763 radiomics features were first computed from the segmented primary cervical tumor depicted on one CT image with the maximal tumor region. Next, a principal component analysis algorithm was applied on the initial feature pool to determine an optimal feature cluster. Then, based on this optimal cluster, the prediction models (i.e., logistic regression or support vector machine) were trained and optimized to generate an image marker to detect LN metastasis. In this study, a retrospective dataset containing 127 cervical cancer patients were established to build and test the model. The model was trained using a leave-one-case-out (LOCO) cross-validation strategy and image marker performance was evaluated using the area under receiver operation characteristic (ROC) curve (AUC).

RESULTS

The results indicate that the SVM based imaging marker achieved an AUC value of 0.841 ± 0.035. When setting an operating threshold of 0.5 on model-generated prediction scores, the imaging marker yielded a positive and negative predictive value (PPV and NPV) of 0.762 and 0.765 respectively, while the total accuracy is 76.4%.

CONCLUSIONS

This study initially verified the feasibility of utilizing CT image and radiomics technology to develop a low-cost image marker to detect LN metastasis for assisting stratification of cervical cancer patients.

Development and Assessment of a New Global Mammographic Image Feature Analysis Scheme to Predict Likelihood of Malignant Cases

This study aims to develop and evaluate a new computer-aided diagnosis (CADx) scheme based on analysis of global mammographic image features to predict likelihood of cases being malignant. An image dataset involving 1,959 cases was retrospectively assembled. Suspicious lesions were detected and biopsied in each case. Among them, 737 cases are malignant and 1,222 are benign. Each case includes four mammograms of craniocaudal and mediolateral oblique view of left and right breasts. CADx scheme is applied to pre-process mammograms, generate two image maps in frequency domain using discrete cosine transform and fast Fourier transform, compute bilateral image feature differences from left and right breasts, and apply a support vector machine (SVM) to predict likelihood of the case being malignant. Three sub-groups of image features were computed from the original mammograms and two transformation maps. Four SVMs using three sub-groups of image features and fusion of all features were trained and tested using a 10-fold cross-validation method. The computed areas under receiver operating characteristic curves (AUCs) range from 0.85 to 0.91 using image features computed from one of three sub-groups, respectively. By fusion of all image features computed in three sub-groups, the fourth SVM yields a significantly higher performance with AUC = 0.96±0.01 (p<0.01). This study demonstrates feasibility of developing a new global image feature analysis based CADx scheme of mammograms with high performance. By avoiding difficulty and possible errors in breast lesion segmentation, this new CADx approach is more efficient in development and potentially more robust in future application.

Applying a new quantitative image analysis scheme based on global mammographic features to assist diagnosis of breast cancer

BACKGROUND AND OBJECTIVE

This study aims to develop and evaluate a unique global mammographic image feature analysis scheme to predict likelihood of a case depicting the detected suspicious breast mass being malignant for breast cancer.

METHODS

From the entire breast area depicting on the mammograms, 59 features were initially computed to characterize the breast tissue properties at both spatial and frequency domain. Given that each case consists of two cranio-caudal and two medio-lateral oblique view images of left and right breasts, two feature pools were built, which contain the computed features from either two positive images of one breast or all the four images of two breasts. Next, for each feature pool, a particle swarm optimization (PSO) method was applied to determine the optimal feature cluster followed by training a support vector machine (SVM) classifier to generate a final score for predicting likelihood of the case being malignant. To test the scheme, we assembled a dataset involving 275 patients who had biopsy due to the suspicious findings on mammograms. Among them, 134 are malignant and 141 are benign. A ten-fold cross validation method was used to train and test the scheme.

RESULTS

The classification performance levels measured by the areas under ROC curves are 0.79 ± 0.07 and 0.75 ± 0.08 when applying the SVM classifiers trained using image features computed from two-view and four-view images, respectively.

CONCLUSIONS

This study demonstrates feasibility of developing a new global mammographic image feature analysis-based scheme to predict the likelihood of case being malignant without lesion segmentation.

Artificial Intelligence (AI) for the early detection of breast cancer: a scoping review to assess AI's potential in breast screening practice

INTRODUCTION

Various factors are driving interest in the application of artificial intelligence (AI) for breast cancer (BC) detection, but it is unclear whether the evidence warrants large-scale use in population-based screening.

AREAS COVERED

We performed a scoping review, a structured evidence synthesis describing a broad research field, to summarize knowledge on AI evaluated for BC detection and to assess AI's readiness for adoption in BC screening. Studies were predominantly small retrospective studies based on highly selected image datasets that contained a high proportion of cancers (median BC proportion in datasets 26.5%), and used heterogeneous techniques to develop AI models; the range of estimated AUC (area under ROC curve) for AI models was 69.2-97.8% (median AUC 88.2%). We identified various methodologic limitations including use of non-representative imaging data for model training, limited validation in external datasets, potential bias in training data, and few comparative data for AI versus radiologists' interpretation of mammography screening.

EXPERT OPINION

Although contemporary AI models have reported generally good accuracy for BC detection, methodological concerns, and evidence gaps exist that limit translation into clinical BC screening settings. These should be addressed in parallel to advancing AI techniques to render AI transferable to large-scale population-based screening.

Developing global image feature analysis models to predict cancer risk and prognosis

In order to develop precision or personalized medicine, identifying new quantitative imaging markers and building machine learning models to predict cancer risk and prognosis has been attracting broad research interest recently. Most of these research approaches use the similar concepts of the conventional computer-aided detection schemes of medical images, which include steps in detecting and segmenting suspicious regions or tumors, followed by training machine learning models based on the fusion of multiple image features computed from the segmented regions or tumors. However, due to the heterogeneity and boundary fuzziness of the suspicious regions or tumors, segmenting subtle regions is often difficult and unreliable. Additionally, ignoring global and/or background parenchymal tissue characteristics may also be a limitation of the conventional approaches. In our recent studies, we investigated the feasibility of developing new computer-aided schemes implemented with the machine learning models that are trained by global image features to predict cancer risk and prognosis. We trained and tested several models using images obtained from full-field digital mammography, magnetic resonance imaging, and computed tomography of breast, lung, and ovarian cancers. Study results showed that many of these new models yielded higher performance than other approaches used in current clinical practice. Furthermore, the computed global image features also contain complementary information from the features computed from the segmented regions or tumors in predicting cancer prognosis. Therefore, the global image features can be used alone to develop new case-based prediction models or can be added to current tumor-based models to increase their discriminatory power.

Machine learning techniques for breast cancer computer aided diagnosis using different image modalities: A systematic review

BACKGROUND AND OBJECTIVE

The high incidence of breast cancer in women has increased significantly in the recent years. Physician experience of diagnosing and detecting breast cancer can be assisted by using some computerized features extraction and classification algorithms. This paper presents the conduction and results of a systematic review (SR) that aims to investigate the state of the art regarding the computer aided diagnosis/detection (CAD) systems for breast cancer.

METHODS

The SR was conducted using a comprehensive selection of scientific databases as reference sources, allowing access to diverse publications in the field. The scientific databases used are Springer Link (SL), Science Direct (SD), IEEE Xplore Digital Library, and PubMed. Inclusion and exclusion criteria were defined and applied to each retrieved work to select those of interest. From 320 studies retrieved, 154 studies were included. However, the scope of this research is limited to scientific and academic works and excludes commercial interests.

RESULTS

This survey provides a general analysis of the current status of CAD systems according to the used image modalities and the machine learning based classifiers. Potential research studies have been discussed to create a more objective and efficient CAD systems.

Prediction of breast cancer risk using a machine learning approach embedded with a locality preserving projection algorithm

In order to automatically identify a set of effective mammographic image features and build an optimal breast cancer risk stratification model, this study aims to investigate advantages of applying a machine learning approach embedded with a locally preserving projection (LPP) based feature combination and regeneration algorithm to predict short-term breast cancer risk. A dataset involving negative mammograms acquired from 500 women was assembled. This dataset was divided into two age-matched classes of 250 high risk cases in which cancer was detected in the next subsequent mammography screening and 250 low risk cases, which remained negative. First, a computer-aided image processing scheme was applied to segment fibro-glandular tissue depicted on mammograms and initially compute 44 features related to the bilateral asymmetry of mammographic tissue density distribution between left and right breasts. Next, a multi-feature fusion based machine learning classifier was built to predict the risk of cancer detection in the next mammography screening. A leave-one-case-out (LOCO) cross-validation method was applied to train and test the machine learning classifier embedded with a LLP algorithm, which generated a new operational vector with 4 features using a maximal variance approach in each LOCO process. Results showed a 9.7% increase in risk prediction accuracy when using this LPP-embedded machine learning approach. An increased trend of adjusted odds ratios was also detected in which odds ratios increased from 1.0 to 11.2. This study demonstrated that applying the LPP algorithm effectively reduced feature dimensionality, and yielded higher and potentially more robust performance in predicting short-term breast cancer risk.

Involvement of Machine Learning for Breast Cancer Image Classification: A Survey

Breast cancer is one of the largest causes of women's death in the world today. Advance engineering of natural image classification techniques and Artificial Intelligence methods has largely been used for the breast-image classification task. The involvement of digital image classification allows the doctor and the physicians a second opinion, and it saves the doctors' and physicians' time. Despite the various publications on breast image classification, very few review papers are available which provide a detailed description of breast cancer image classification techniques, feature extraction and selection procedures, classification measuring parameterizations, and image classification findings. We have put a special emphasis on the Convolutional Neural Network (CNN) method for breast image classification. Along with the CNN method we have also described the involvement of the conventional Neural Network (NN), Logic Based classifiers such as the Random Forest (RF) algorithm, Support Vector Machines (SVM), Bayesian methods, and a few of the semisupervised and unsupervised methods which have been used for breast image classification.

A Machine Learning-based Surface Electromyography Topography Evaluation for Prognostic Prediction of Functional Restoration Rehabilitation in Chronic Low Back Pain

STUDY DESIGN

A retrospective study.

OBJECTIVE

The aim of this study was to investigate the feasibility and applicability of support vector machine (SVM) algorithm in classifying patients with LBP who would obtain satisfactory or unsatisfactory progress after the functional restoration rehabilitation program.

SUMMARY OF BACKGROUND DATA

Dynamic surface electromyography (SEMG) topography has demonstrated the potential use in predicting the prognosis of functional restoration rehabilitation for patients with low back pain (LBP). However, processing from raw SEMG topography to make prediction is not easy to clinicians.

METHODS

A total of 30 patients with nonspecific LBP were recruited and divided into "responding" and "non-responding" group according to the change of Visual analog pain rating scale and Oswestry Disability Index. Each patient received a 12-week functional restoration rehabilitation program. A normal database was calculated from a control group from 48 healthy participants. Root-mean-square difference (RMSD) was extracted from the recorded dynamic SEMG topography during symmetrical and asymmetrical trunk-movement. SVM and cross-validation were applied to the prediction based on the optimized features selected by the sequential floating forward selection (SFFS) algorithm.

RESULTS

RMSD feature parameters following rehabilitation in the "responding" group showed a significant difference (P < 0.05) with the one in the "nonresponding" group. The SVM classifier with Quadratic kernel based on SFFS-selected features showed the best prediction performance (accuracy: 96.67%, sensitivity: 100%, specificity: 93.75%, average area under curve [AUC]: 0.8925) comparing with linear kernel (accuracy: 80.00%, sensitivity: 85.71%, specificity: 75.00%, average AUC: 0.7825), polynomial kernel (accuracy: 93.33%, sensitivity: 92.86%, specificity: 93.75%, average AUC: 0.9675), and radial basis function (RBF) kernel (accuracy: 86.67%, sensitivity: 85.71%, specificity: 87.50%, average AUC: 0.7900).

CONCLUSION

The use of SVM-based classifier of SEMG topography can be applied to identify the patient responding to functional restoration rehabilitation, which will help the healthcare worker to improve the efficiency of LBP rehabilitation.

LEVEL OF EVIDENCE

3.

Neighborhood Structural Similarity Mapping for the Classification of Masses in Mammograms

In this paper, two novel feature extraction methods, using neighborhood structural similarity (NSS), are proposed for the characterization of mammographic masses as benign or malignant. Since gray-level distribution of pixels is different in benign and malignant masses, more regular and homogeneous patterns are visible in benign masses compared to malignant masses; the proposed method exploits the similarity between neighboring regions of masses by designing two new features, namely, NSS-I and NSS-II, which capture global similarity at different scales. Complementary to these global features, uniform local binary patterns are computed to enhance the classification efficiency by combining with the proposed features. The performance of the features are evaluated using the images from the mini-mammographic image analysis society (mini-MIAS) and digital database for screening mammography (DDSM) databases, where a tenfold cross-validation technique is incorporated with Fisher linear discriminant analysis, after selecting the optimal set of features using stepwise logistic regression method. The best area under the receiver operating characteristic curve of 0.98 with an accuracy of is achieved with the mini-MIAS database, while the same for the DDSM database is 0.93 with accuracy .

A new approach to develop computer-aided diagnosis scheme of breast mass classification using deep learning technology

PURPOSE

To develop and test a deep learning based computer-aided diagnosis (CAD) scheme of mammograms for classifying between malignant and benign masses.

METHODS

An image dataset involving 560 regions of interest (ROIs) extracted from digital mammograms was used. After down-sampling each ROI from 512×512 to 64×64 pixel size, we applied an 8 layer deep learning network that involves 3 pairs of convolution-max-pooling layers for automatic feature extraction and a multiple layer perceptron (MLP) classifier for feature categorization to process ROIs. The 3 pairs of convolution layers contain 20, 10, and 5 feature maps, respectively. Each convolution layer is connected with a max-pooling layer to improve the feature robustness. The output of the sixth layer is fully connected with a MLP classifier, which is composed of one hidden layer and one logistic regression layer. The network then generates a classification score to predict the likelihood of ROI depicting a malignant mass. A four-fold cross validation method was applied to train and test this deep learning network.

RESULTS

The results revealed that this CAD scheme yields an area under the receiver operation characteristic curve (AUC) of 0.696±0.044, 0.802±0.037, 0.836±0.036, and 0.822±0.035 for fold 1 to 4 testing datasets, respectively. The overall AUC of the entire dataset is 0.790±0.019.

CONCLUSIONS

This study demonstrates the feasibility of applying a deep learning based CAD scheme to classify between malignant and benign breast masses without a lesion segmentation, image feature computation and selection process.

Computer-aided classification of mammographic masses using visually sensitive image features

PURPOSE

To develop a new computer-aided diagnosis (CAD) scheme that computes visually sensitive image features routinely used by radiologists to develop a machine learning classifier and distinguish between the malignant and benign breast masses detected from digital mammograms.

METHODS

An image dataset including 301 breast masses was retrospectively selected. From each segmented mass region, we computed image features that mimic five categories of visually sensitive features routinely used by radiologists in reading mammograms. We then selected five optimal features in the five feature categories and applied logistic regression models for classification. A new CAD interface was also designed to show lesion segmentation, computed feature values and classification score.

RESULTS

Areas under ROC curves (AUC) were 0.786±0.026 and 0.758±0.027 when to classify mass regions depicting on two view images, respectively. By fusing classification scores computed from two regions, AUC increased to 0.806±0.025.

CONCLUSION

This study demonstrated a new approach to develop CAD scheme based on 5 visually sensitive image features. Combining with a "visual aid" interface, CAD results may be much more easily explainable to the observers and increase their confidence to consider CAD generated classification results than using other conventional CAD approaches, which involve many complicated and visually insensitive texture features.

Computer-Extracted Texture Features to Distinguish Cerebral Radionecrosis from Recurrent Brain Tumors on Multiparametric MRI: A Feasibility Study

BACKGROUND AND PURPOSE

Despite availability of advanced imaging, distinguishing radiation necrosis from recurrent brain tumors noninvasively is a big challenge in neuro-oncology. Our aim was to determine the feasibility of radiomic (computer-extracted texture) features in differentiating radiation necrosis from recurrent brain tumors on routine MR imaging (gadolinium T1WI, T2WI, FLAIR).

MATERIALS AND METHODS

A retrospective study of brain tumor MR imaging performed 9 months (or later) post-radiochemotherapy was performed from 2 institutions. Fifty-eight patient studies were analyzed, consisting of a training (n = 43) cohort from one institution and an independent test (n = 15) cohort from another, with surgical histologic findings confirmed by an experienced neuropathologist at the respective institutions. Brain lesions on MR imaging were manually annotated by an expert neuroradiologist. A set of radiomic features was extracted for every lesion on each MR imaging sequence: gadolinium T1WI, T2WI, and FLAIR. Feature selection was used to identify the top 5 most discriminating features for every MR imaging sequence on the training cohort. These features were then evaluated on the test cohort by a support vector machine classifier. The classification performance was compared against diagnostic reads by 2 expert neuroradiologists who had access to the same MR imaging sequences (gadolinium T1WI, T2WI, and FLAIR) as the classifier.

RESULTS

On the training cohort, the area under the receiver operating characteristic curve was highest for FLAIR with 0.79; 95% CI, 0.77-0.81 for primary (n = 22); and 0.79, 95% CI, 0.75-0.83 for metastatic subgroups (n = 21). Of the 15 studies in the holdout cohort, the support vector machine classifier identified 12 of 15 studies correctly, while neuroradiologist 1 diagnosed 7 of 15 and neuroradiologist 2 diagnosed 8 of 15 studies correctly, respectively.

CONCLUSIONS

Our preliminary results suggest that radiomic features may provide complementary diagnostic information on routine MR imaging sequences that may improve the distinction of radiation necrosis from recurrence for both primary and metastatic brain tumors.

Assessment of performance and reproducibility of applying a content-based image retrieval scheme for classification of breast lesions

PURPOSE

To develop a new computer-aided diagnosis (CAD) scheme using a content-based image retrieval (CBIR) approach for classification between the malignant and benign breast lesions depicted on the digital mammograms and assess CAD performance and reproducibility.

METHODS

An image dataset including 820 regions of interest (ROIs) was used. Among them, 431 ROIs depict malignant lesions and 389 depict benign lesions. After applying an image preprocessing process to define the lesion center, two image features were computed from each ROI. The first feature is an average pixel value of a mapped region generated using a watershed algorithm. The second feature is an average pixel value difference between a ROI's center region and the rest of the image. A two-step CBIR approach uses these two features sequentially to search for ten most similar reference ROIs for each queried ROI. A similarity based classification score was then computed to predict the likelihood of the queried ROI depicting a malignant lesion. To assess the reproducibility of the CAD scheme, we selected another independent testing dataset of 100 ROIs. For each ROI in the testing dataset, we added four randomly queried lesion center pixels and examined the variation of the classification scores.

RESULTS

The area under the ROC curve (AUC) = 0.962 ± 0.006 was obtained when applying a leave-one-out validation method to 820 ROIs. Using the independent testing dataset, the initial AUC value was 0.832 ± 0.040, and using the median classification score of each ROI with five queried seeds, AUC value increased to 0.878 ± 0.035.

CONCLUSIONS

The authors demonstrated that (1) a simple and efficient CBIR scheme using two lesion density distribution related features achieved high performance in classifying breast lesions without actual lesion segmentation and (2) similar to the conventional CAD schemes using global optimization approaches, improving reproducibility is also one of the challenges in developing CAD schemes using a CBIR based regional optimization approach.

An automated confirmatory system for analysis of mammograms

This paper presents an integrated system for the automatic analysis of mammograms to assist radiologists in confirming their diagnosis in mammography screening. The proposed automated confirmatory system (ACS) can process a digitalized mammogram online, and generates a high quality filtered segmentation of an image for biological interpretation and a texture-feature based diagnosis. We use a serial of image pre-processing and segmentation techniques, including 2D median filtering, seeded region growing (SRG) algorithm, image contrast enhancement, to remove noise, delete radiopaque artifacts and eliminate the projection of the pectoral muscle from a digitalized mammogram. We also develop an entire-image texture-feature based classification method, by combining a Rough-set approach to extract five fundamental texture features from images, and then an Artificial Neural Network technique to classify a mammogram as: normal; indicating the presence of a benign lump; or representing a malignant tumor. Here, 222 random images from the Mammographic Image Analysis Society (MIAS) database are used for the offline ACS training. Once the system is tuned and trained, it is ready for the automated use for the analysis and diagnosis of new mammograms. To test the trained system, a separate set of 100 random images from the MIAS and another set of 100 random images from the independent BancoWeb database are selected. The proposed ACS is shown to be successful in confirming diagnosis of mammograms from the two independent databases.

A new and fast image feature selection method for developing an optimal mammographic mass detection scheme

PURPOSE

Selecting optimal features from a large image feature pool remains a major challenge in developing computer-aided detection (CAD) schemes of medical images. The objective of this study is to investigate a new approach to significantly improve efficacy of image feature selection and classifier optimization in developing a CAD scheme of mammographic masses.

METHODS

An image dataset including 1600 regions of interest (ROIs) in which 800 are positive (depicting malignant masses) and 800 are negative (depicting CAD-generated false positive regions) was used in this study. After segmentation of each suspicious lesion by a multilayer topographic region growth algorithm, 271 features were computed in different feature categories including shape, texture, contrast, isodensity, spiculation, local topological features, as well as the features related to the presence and location of fat and calcifications. Besides computing features from the original images, the authors also computed new texture features from the dilated lesion segments. In order to select optimal features from this initial feature pool and build a highly performing classifier, the authors examined and compared four feature selection methods to optimize an artificial neural network (ANN) based classifier, namely: (1) Phased Searching with NEAT in a Time-Scaled Framework, (2) A sequential floating forward selection (SFFS) method, (3) A genetic algorithm (GA), and (4) A sequential forward selection (SFS) method. Performances of the four approaches were assessed using a tenfold cross validation method.

RESULTS

Among these four methods, SFFS has highest efficacy, which takes 3%-5% of computational time as compared to GA approach, and yields the highest performance level with the area under a receiver operating characteristic curve (AUC) = 0.864 ± 0.034. The results also demonstrated that except using GA, including the new texture features computed from the dilated mass segments improved the AUC results of the ANNs optimized using other three feature selection methods. In addition, among 271 features, the shape, local morphological features, fat and calcification based features were the most frequently selected features to build ANNs.

CONCLUSIONS

Although conventional GA is a powerful tool in optimizing classifiers used in CAD schemes of medical images, it is very computationally intensive. This study demonstrated that using a new SFFS based approach enabled to significantly improve efficacy of image feature selection for developing CAD schemes.

Pharmacokinetics and tissue distribution model of cabozantinib in rat determined by UPLC-MS/MS

Cabozantinib (XL184) is a novel small molecule inhibitor of receptor tyrosine kinases (RTKs) targeted at mesenchymal-epithelial transition factor (MET). In order to study the pharmacokinetics and tissue distribution in rat, a specific ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) method was developed with midazolam as internal standard. The calibration curves in plasma and tissues were linear in the range of 5-5000ng/mL (r(2)>0.99). The recoveries were better than 80.4% and matrix effects ranged from 96.9% to 105.1%. Then, the developed UPLC-MS/MS method was applied to determine the concentration of XL184 in blood and tissues. The pharmacokinetics of four different dosages (iv 5, 10mg/kg and ig 15, 30mg/kg) revealed that XL184 was eliminated slowly, the t1/2 was longer than 10h and the absolute bioavailability was 25.6±8.3%. The concentration distribution of XL184 in tissues was liver>lung>kidney>spleen>heart. Based on the concentration-time of XL184 in tissues, a BP-ANN distribution model was developed with good performance, and can be used to predict the concentration of XL184 in tissues.

Optimization of Network Topology in Computer-Aided Detection Schemes Using Phased Searching with NEAT in a Time-Scaled Framework

In the field of computer-aided mammographic mass detection, many different features and classifiers have been tested. Frequently, the relevant features and optimal topology for the artificial neural network (ANN)-based approaches at the classification stage are unknown, and thus determined by trial-and-error experiments. In this study, we analyzed a classifier that evolves ANNs using genetic algorithms (GAs), which combines feature selection with the learning task. The classifier named "Phased Searching with NEAT in a Time-Scaled Framework" was analyzed using a dataset with 800 malignant and 800 normal tissue regions in a 10-fold cross-validation framework. The classification performance measured by the area under a receiver operating characteristic (ROC) curve was 0.856 ± 0.029. The result was also compared with four other well-established classifiers that include fixed-topology ANNs, support vector machines (SVMs), linear discriminant analysis (LDA), and bagged decision trees. The results show that Phased Searching outperformed the LDA and bagged decision tree classifiers, and was only significantly outperformed by SVM. Furthermore, the Phased Searching method required fewer features and discarded superfluous structure or topology, thus incurring a lower feature computational and training and validation time requirement. Analyses performed on the network complexities evolved by Phased Searching indicate that it can evolve optimal network topologies based on its complexification and simplification parameter selection process. From the results, the study also concluded that the three classifiers - SVM, fixed-topology ANN, and Phased Searching with NeuroEvolution of Augmenting Topologies (NEAT) in a Time-Scaled Framework - are performing comparably well in our mammographic mass detection scheme.