Bayesian neural network approach for determining the risk of re-intervention after endovascular aortic aneurysm repair

Applying Artificial Intelligence to Predict Complications After Endovascular Aneurysm Repair

Objective: Complications after Endovascular Aneurysm Repair (EVAR) can be fatal. Patient follow-up for surveillance imaging is becoming more challenging as fewer patients are seen, particularly after the first year. The aim of this study was to develop an artificial intelligence model to predict the complication probability of individual patients to better identify those needing more intensive post-operative surveillance. Methods: Pre-operative CTA 3D reconstruction images of AAA from 273 patients who underwent EVAR from 2011-2020 were collected. Of these, 48 patients had post-operative complications including endoleak, AAA rupture, graft limb occlusion, renal artery occlusion, and neck dilation. A deep convolutional neural network model (VascAI©) was developed which utilized pre-operative 3D CT images to predict risk of complications after EVAR. The model was built with TensorFlow software and run on the Google Colab Platform. An initial training subset of 40 randomly selected patients with complications and 189 without were used to train the AI model while the remaining 8 positive and 36 negative cases tested its performance and prediction accuracy. Data down-sampling was used to alleviate data imbalance and data augmentation methodology to further boost model performance. Results: Successful training was completed on the 229 cases in the training set and then applied to predict the complication probability of each individual in the held-out performance testing cases. The model provided a complication sensitivity of 100% and identified all the patients who later developed complications after EVAR. Of 36 patients without complications, 16 (44%) were falsely predicted to develop complications. The results therefore demonstrated excellent sensitivity for identifying patients who would benefit from more stringent surveillance and decrease the frequency of surveillance in 56% of patients unlike to develop complications. Conclusion: AI models can be developed to predict the risk of post-operative complications with high accuracy. Compared to existing methods, the model developed in this study did not require any expert-annotated data but only the AAA CTA images as inputs. This model can play an assistive role in identifying patients at high risk for post-EVAR complications and the need for greater compliance in surveillance.

Machine learning to predict outcomes following endovascular abdominal aortic aneurysm repair

BACKGROUND

Endovascular aneurysm repair (EVAR) for abdominal aortic aneurysm (AAA) carries important perioperative risks; however, there are no widely used outcome prediction tools. The aim of this study was to apply machine learning (ML) to develop automated algorithms that predict 1-year mortality following EVAR.

METHODS

The Vascular Quality Initiative database was used to identify patients who underwent elective EVAR for infrarenal AAA between 2003 and 2023. Input features included 47 preoperative demographic/clinical variables. The primary outcome was 1-year all-cause mortality. Data were split into training (70 per cent) and test (30 per cent) sets. Using 10-fold cross-validation, 6 ML models were trained using preoperative features with logistic regression as the baseline comparator. The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). Model robustness was evaluated with calibration plot and Brier score.

RESULTS

Some 63 655 patients were included. One-year mortality occurred in 3122 (4.9 per cent) patients. The best performing prediction model for 1-year mortality was XGBoost, achieving an AUROC (95 per cent c.i.) of 0.96 (0.95-0.97). Comparatively, logistic regression had an AUROC (95 per cent c.i.) of 0.69 (0.68-0.71). The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.04. The top 3 predictive features in the algorithm were 1) unfit for open AAA repair, 2) functional status, and 3) preoperative dialysis.

CONCLUSIONS

In this data set, machine learning was able to predict 1-year mortality following EVAR using preoperative data and outperformed standard logistic regression models.

GabROP: Gabor Wavelets-Based CAD for Retinopathy of Prematurity Diagnosis via Convolutional Neural Networks

One of the most serious and dangerous ocular problems in premature infants is retinopathy of prematurity (ROP), a proliferative vascular disease. Ophthalmologists can use automatic computer-assisted diagnostic (CAD) tools to help them make a safe, accurate, and low-cost diagnosis of ROP. All previous CAD tools for ROP diagnosis use the original fundus images. Unfortunately, learning the discriminative representation from ROP-related fundus images is difficult. Textural analysis techniques, such as Gabor wavelets (GW), can demonstrate significant texture information that can help artificial intelligence (AI) based models to improve diagnostic accuracy. In this paper, an effective and automated CAD tool, namely GabROP, based on GW and multiple deep learning (DL) models is proposed. Initially, GabROP analyzes fundus images using GW and generates several sets of GW images. Next, these sets of images are used to train three convolutional neural networks (CNNs) models independently. Additionally, the actual fundus pictures are used to build these networks. Using the discrete wavelet transform (DWT), texture features retrieved from every CNN trained with various sets of GW images are combined to create a textural-spectral-temporal demonstration. Afterward, for each CNN, these features are concatenated with spatial deep features obtained from the original fundus images. Finally, the previous concatenated features of all three CNN are incorporated using the discrete cosine transform (DCT) to lessen the size of features caused by the fusion process. The outcomes of GabROP show that it is accurate and efficient for ophthalmologists. Additionally, the effectiveness of GabROP is compared to recently developed ROP diagnostic techniques. Due to GabROP's superior performance compared to competing tools, ophthalmologists may be able to identify ROP more reliably and precisely, which could result in a reduction in diagnostic effort and examination time.

A wavelet-based deep learning pipeline for efficient COVID-19 diagnosis via CT slices

The quick diagnosis of the novel coronavirus (COVID-19) disease is vital to prevent its propagation and improve therapeutic outcomes. Computed tomography (CT) is believed to be an effective tool for diagnosing COVID-19, however, the CT scan contains hundreds of slices that are complex to be analyzed and could cause delays in diagnosis. Artificial intelligence (AI) especially deep learning (DL), could facilitate and speed up COVID-19 diagnosis from such scans. Several studies employed DL approaches based on 2D CT images from a single view, nevertheless, 3D multiview CT slices demonstrated an excellent ability to enhance the efficiency of COVID-19 diagnosis. The majority of DL-based studies utilized the spatial information of the original CT images to train their models, though, using spectral–temporal information could improve the detection of COVID-19. This article proposes a DL-based pipeline called CoviWavNet for the automatic diagnosis of COVID-19. CoviWavNet uses a 3D multiview dataset called OMNIAHCOV. Initially, it analyzes the CT slices using multilevel discrete wavelet decomposition (DWT) and then uses the heatmaps of the approximation levels to train three ResNet CNN models. These ResNets use the spectral–temporal information of such images to perform classification. Subsequently, it investigates whether the combination of spatial information with spectral–temporal information could improve the diagnostic accuracy of COVID-19. For this purpose, it extracts deep spectral–temporal features from such ResNets using transfer learning and integrates them with deep spatial features extracted from the same ResNets trained with the original CT slices. Then, it utilizes a feature selection step to reduce the dimension of such integrated features and use them as inputs to three support vector machine (SVM) classifiers. To further validate the performance of CoviWavNet, a publicly available benchmark dataset called SARS-COV-2-CT-Scan is employed. The results of CoviWavNet have demonstrated that using the spectral–temporal information of the DWT heatmap images to train the ResNets is superior to utilizing the spatial information of the original CT images. Furthermore, integrating deep spectral–temporal features with deep spatial features has enhanced the classification accuracy of the three SVM classifiers reaching a final accuracy of 99.33% and 99.7% for the OMNIAHCOV and SARS-COV-2-CT-Scan datasets respectively. These accuracies verify the outstanding performance of CoviWavNet compared to other related studies. Thus, CoviWavNet can help radiologists in the rapid and accurate diagnosis of COVID-19 diagnosis.

Fully automatic volume segmentation using deep learning approaches to assess aneurysmal sac evolution after infra-renal endovascular aortic repair

OBJECTIVE

Endovascular aortic repair (EVAR) surveillance relies on serial measurements of maximal diameter despite significant inter- and intra-observer variability. Volumetric measurements are more sensitive but general use is hampered by the time required for their implementation. An innovative fully automated software (PRAEVAorta® from Nurea), using artificial intelligence (AI), previously demonstrated fast and robust detection of infra-renal abdominal aortic aneurysm's (AAA) characteristics on pre-operative imaging. This study aimed to assess the robustness of these data on post-EVAR computed tomography (CT) scans.

METHODS

Comparison was made between fully automatic and semi-automatic segmentation manually corrected by a senior surgeon on a dataset of 48 patients (48 early post-EVAR CT scans with 6466 slices, and a total of 101 follow-up CT scans with 13708 slices).

RESULTS

The analyses confirmed an excellent correlation of post-EVAR volumes and surfaces, as well as, proximal neck and maximum aneurysm diameters measured with the fully automatic and manually corrected segmentation methods (Pearson's coefficient correlation >.99, p<.0001). Comparison between the fully automatic and manually corrected segmentation method revealed a mean Dice Similarity Coefficient of 0.950±0.015, Jaccard index of 0.906±0.028, Sensitivity of 0.929±0.028, Specificity of 0.965±0.016, Volumetric Similarity (VS) of 0.973±0.018 and mean Hausdorff Distance/slice of 8.7±10.8mm. The mean VS reached 0.873±0.100 for the lumen and 0.903±0.091 for the thrombus. The segmentation time was 9 times faster with the fully automatic method (2.5 vs 22 min/patient with the manually corrected method; p<.0001). Preliminary analysis also demonstrated that a diameter increase of 2mm can actually represent >5% volume increase.

CONCLUSION

PRAEVAorta® enables a fast, reproducible, and fully automated analysis of post-EVAR AAA sac and neck characteristics, with comparison between different time points. It could become a crucial adjunct for EVAR follow-up through early detection of sac evolution, which may reduce the risk of secondary rupture.

ECG-BiCoNet: An ECG-based pipeline for COVID-19 diagnosis using Bi-Layers of deep features integration

The accurate and speedy detection of COVID-19 is essential to avert the fast propagation of the virus, alleviate lockdown constraints and diminish the burden on health organizations. Currently, the methods used to diagnose COVID-19 have several limitations, thus new techniques need to be investigated to improve the diagnosis and overcome these limitations. Taking into consideration the great benefits of electrocardiogram (ECG) applications, this paper proposes a new pipeline called ECG-BiCoNet to investigate the potential of using ECG data for diagnosing COVID-19. ECG-BiCoNet employs five deep learning models of distinct structural design. ECG-BiCoNet extracts two levels of features from two different layers of each deep learning technique. Features mined from higher layers are fused using discrete wavelet transform and then integrated with lower-layers features. Afterward, a feature selection approach is utilized. Finally, an ensemble classification system is built to merge predictions of three machine learning classifiers. ECG-BiCoNet accomplishes two classification categories, binary and multiclass. The results of ECG-BiCoNet present a promising COVID-19 performance with an accuracy of 98.8% and 91.73% for binary and multiclass classification categories. These results verify that ECG data may be used to diagnose COVID-19 which can help clinicians in the automatic diagnosis and overcome limitations of manual diagnosis.

A computer-aided diagnostic framework for coronavirus diagnosis using texture-based radiomics images

The accurate and rapid detection of the novel coronavirus infection, coronavirus is very important to prevent the fast spread of such disease. Thus, reducing negative effects that influenced many industrial sectors, especially healthcare. Artificial intelligence techniques in particular deep learning could help in the fast and precise diagnosis of coronavirus from computed tomography images. Most artificial intelligence-based studies used the original computed tomography images to build their models; however, the integration of texture-based radiomics images and deep learning techniques could improve the diagnostic accuracy of the novel coronavirus diseases. This study proposes a computer-assisted diagnostic framework based on multiple deep learning and texture-based radiomics approaches. It first trains three Residual Networks (ResNets) deep learning techniques with two texture-based radiomics images including discrete wavelet transform and gray-level covariance matrix instead of the original computed tomography images. Then, it fuses the texture-based radiomics deep features sets extracted from each using discrete cosine transform. Thereafter, it further combines the fused texture-based radiomics deep features obtained from the three convolutional neural networks. Finally, three support vector machine classifiers are utilized for the classification procedure. The proposed method is validated experimentally on the benchmark severe respiratory syndrome coronavirus 2 computed tomography image dataset. The accuracies attained indicate that using texture-based radiomics (gray-level covariance matrix, discrete wavelet transform) images for training the ResNet-18 (83.22%, 74.9%), ResNet-50 (80.94%, 78.39%), and ResNet-101 (80.54%, 77.99%) is better than using the original computed tomography images (70.34%, 76.51%, and 73.42%) for ResNet-18, ResNet-50, and ResNet-101, respectively. Furthermore, the sensitivity, specificity, accuracy, precision, and F1-score achieved using the proposed computer-assisted diagnostic after the two fusion steps are 99.47%, 99.72%, 99.60%, 99.72%, and 99.60% which proves that combining texture-based radiomics deep features obtained from the three ResNets has boosted its performance. Thus, fusing multiple texture-based radiomics deep features mined from several convolutional neural networks is better than using only one type of radiomics approach and a single convolutional neural network. The performance of the proposed computer-assisted diagnostic framework allows it to be used by radiologists in attaining fast and accurate diagnosis.

A deep learning-based diagnostic tool for identifying various diseases via facial images

With the current health crisis caused by the COVID-19 pandemic, patients have become more anxious about infection, so they prefer not to have direct contact with doctors or clinicians. Lately, medical scientists have confirmed that several diseases exhibit corresponding specific features on the face the face. Recent studies have indicated that computer-aided facial diagnosis can be a promising tool for the automatic diagnosis and screening of diseases from facial images. However, few of these studies used deep learning (DL) techniques. Most of them focused on detecting a single disease, using handcrafted feature extraction methods and conventional machine learning techniques based on individual classifiers trained on small and private datasets using images taken from a controlled environment. This study proposes a novel computer-aided facial diagnosis system called FaceDisNet that uses a new public dataset based on images taken from an unconstrained environment and could be employed for forthcoming comparisons. It detects single and multiple diseases. FaceDisNet is constructed by integrating several spatial deep features from convolutional neural networks of various architectures. It does not depend only on spatial features but also extracts spatial-spectral features. FaceDisNet searches for the fused spatial-spectral feature set that has the greatest impact on the classification. It employs two feature selection techniques to reduce the large dimension of features resulting from feature fusion. Finally, it builds an ensemble classifier based on stacking to perform classification. The performance of FaceDisNet verifies its ability to diagnose single and multiple diseases. FaceDisNet achieved a maximum accuracy of 98.57% and 98% after the ensemble classification and feature selection steps for binary and multiclass classification categories. These results prove that FaceDisNet is a reliable tool and could be employed to avoid the difficulties and complications of manual diagnosis. Also, it can help physicians achieve accurate diagnoses without the need for physical contact with the patients.

DIAROP: Automated Deep Learning-Based Diagnostic Tool for Retinopathy of Prematurity

Retinopathy of Prematurity (ROP) affects preterm neonates and could cause blindness. Deep Learning (DL) can assist ophthalmologists in the diagnosis of ROP. This paper proposes an automated and reliable diagnostic tool based on DL techniques called DIAROP to support the ophthalmologic diagnosis of ROP. It extracts significant features by first obtaining spatial features from the four Convolution Neural Networks (CNNs) DL techniques using transfer learning and then applying Fast Walsh Hadamard Transform (FWHT) to integrate these features. Moreover, DIAROP explores the best-integrated features extracted from the CNNs that influence its diagnostic capability. The results of DIAROP indicate that DIAROP achieved an accuracy of 93.2% and an area under receiving operating characteristic curve (AUC) of 0.98. Furthermore, DIAROP performance is compared with recent ROP diagnostic tools. Its promising performance shows that DIAROP may assist the ophthalmologic diagnosis of ROP.

Intelligent Dermatologist Tool for Classifying Multiple Skin Cancer Subtypes by Incorporating Manifold Radiomics Features Categories

The rates of skin cancer (SC) are rising every year and becoming a critical health issue worldwide. SC's early and accurate diagnosis is the key procedure to reduce these rates and improve survivability. However, the manual diagnosis is exhausting, complicated, expensive, prone to diagnostic error, and highly dependent on the dermatologist's experience and abilities. Thus, there is a vital need to create automated dermatologist tools that are capable of accurately classifying SC subclasses. Recently, artificial intelligence (AI) techniques including machine learning (ML) and deep learning (DL) have verified the success of computer-assisted dermatologist tools in the automatic diagnosis and detection of SC diseases. Previous AI-based dermatologist tools are based on features which are either high-level features based on DL methods or low-level features based on handcrafted operations. Most of them were constructed for binary classification of SC. This study proposes an intelligent dermatologist tool to accurately diagnose multiple skin lesions automatically. This tool incorporates manifold radiomics features categories involving high-level features such as ResNet-50, DenseNet-201, and DarkNet-53 and low-level features including discrete wavelet transform (DWT) and local binary pattern (LBP). The results of the proposed intelligent tool prove that merging manifold features of different categories has a high influence on the classification accuracy. Moreover, these results are superior to those obtained by other related AI-based dermatologist tools. Therefore, the proposed intelligent tool can be used by dermatologists to help them in the accurate diagnosis of the SC subcategory. It can also overcome manual diagnosis limitations, reduce the rates of infection, and enhance survival rates.

MB-AI-His: Histopathological Diagnosis of Pediatric Medulloblastoma and its Subtypes via AI

Medulloblastoma (MB) is a dangerous malignant pediatric brain tumor that could lead to death. It is considered the most common pediatric cancerous brain tumor. Precise and timely diagnosis of pediatric MB and its four subtypes (defined by the World Health Organization (WHO)) is essential to decide the appropriate follow-up plan and suitable treatments to prevent its progression and reduce mortality rates. Histopathology is the gold standard modality for the diagnosis of MB and its subtypes, but manual diagnosis via a pathologist is very complicated, needs excessive time, and is subjective to the pathologists' expertise and skills, which may lead to variability in the diagnosis or misdiagnosis. The main purpose of the paper is to propose a time-efficient and reliable computer-aided diagnosis (CADx), namely MB-AI-His, for the automatic diagnosis of pediatric MB and its subtypes from histopathological images. The main challenge in this work is the lack of datasets available for the diagnosis of pediatric MB and its four subtypes and the limited related work. Related studies are based on either textural analysis or deep learning (DL) feature extraction methods. These studies used individual features to perform the classification task. However, MB-AI-His combines the benefits of DL techniques and textural analysis feature extraction methods through a cascaded manner. First, it uses three DL convolutional neural networks (CNNs), including DenseNet-201, MobileNet, and ResNet-50 CNNs to extract spatial DL features. Next, it extracts time-frequency features from the spatial DL features based on the discrete wavelet transform (DWT), which is a textural analysis method. Finally, MB-AI-His fuses the three spatial-time-frequency features generated from the three CNNs and DWT using the discrete cosine transform (DCT) and principal component analysis (PCA) to produce a time-efficient CADx system. MB-AI-His merges the privileges of different CNN architectures. MB-AI-His has a binary classification level for classifying among normal and abnormal MB images, and a multi-classification level to classify among the four subtypes of MB. The results of MB-AI-His show that it is accurate and reliable for both the binary and multi-class classification levels. It is also a time-efficient system as both the PCA and DCT methods have efficiently reduced the training execution time. The performance of MB-AI-His is compared with related CADx systems, and the comparison verified the powerfulness of MB-AI-His and its outperforming results. Therefore, it can support pathologists in the accurate and reliable diagnosis of MB and its subtypes from histopathological images. It can also reduce the time and cost of the diagnosis procedure which will correspondingly lead to lower death rates.

Histo-CADx: duo cascaded fusion stages for breast cancer diagnosis from histopathological images

Breast cancer (BC) is one of the most common types of cancer that affects females worldwide. It may lead to irreversible complications and even death due to late diagnosis and treatment. The pathological analysis is considered the gold standard for BC detection, but it is a challenging task. Automatic diagnosis of BC could reduce death rates, by creating a computer aided diagnosis (CADx) system capable of accurately identifying BC at an early stage and decreasing the time consumed by pathologists during examinations. This paper proposes a novel CADx system named Histo-CADx for the automatic diagnosis of BC. Most related studies were based on individual deep learning methods. Also, studies did not examine the influence of fusing features from multiple CNNs and handcrafted features. In addition, related studies did not investigate the best combination of fused features that influence the performance of the CADx. Therefore, Histo-CADx is based on two stages of fusion. The first fusion stage involves the investigation of the impact of fusing several deep learning (DL) techniques with handcrafted feature extraction methods using the auto-encoder DL method. This stage also examines and searches for a suitable set of fused features that could improve the performance of Histo-CADx. The second fusion stage constructs a multiple classifier system (MCS) for fusing outputs from three classifiers, to further improve the accuracy of the proposed Histo-CADx. The performance of Histo-CADx is evaluated using two public datasets; specifically, the BreakHis and the ICIAR 2018 datasets. The results from the analysis of both datasets verified that the two fusion stages of Histo-CADx successfully improved the accuracy of the CADx compared to CADx constructed with individual features. Furthermore, using the auto-encoder for the fusion process has reduced the computation cost of the system. Moreover, the results after the two fusion stages confirmed that Histo-CADx is reliable and has the capacity of classifying BC more accurately compared to other latest studies. Consequently, it can be used by pathologists to help them in the accurate diagnosis of BC. In addition, it can decrease the time and effort needed by medical experts during the examination.

Artificial intelligence in abdominal aortic aneurysm

OBJECTIVE

Abdominal aortic aneurysm (AAA) is a life-threatening disease, and the only curative treatment relies on open or endovascular repair. The decision to treat relies on the evaluation of the risk of AAA growth and rupture, which can be difficult to assess in practice. Artificial intelligence (AI) has revealed new insights into the management of cardiovascular diseases, but its application in AAA has so far been poorly described. The aim of this review was to summarize the current knowledge on the potential applications of AI in patients with AAA.

METHODS

A comprehensive literature review was performed. The MEDLINE database was searched according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The search strategy used a combination of keywords and included studies using AI in patients with AAA published between May 2019 and January 2000. Two authors independently screened titles and abstracts and performed data extraction. The search of published literature identified 34 studies with distinct methodologies, aims, and study designs.

RESULTS

AI was used in patients with AAA to improve image segmentation and for quantitative analysis and characterization of AAA morphology, geometry, and fluid dynamics. AI allowed computation of large data sets to identify patterns that may be predictive of AAA growth and rupture. Several predictive and prognostic programs were also developed to assess patients' postoperative outcomes, including mortality and complications after endovascular aneurysm repair.

CONCLUSIONS

AI represents a useful tool in the interpretation and analysis of AAA imaging by enabling automatic quantitative measurements and morphologic characterization. It could be used to help surgeons in preoperative planning. AI-driven data management may lead to the development of computational programs for the prediction of AAA evolution and risk of rupture as well as postoperative outcomes. AI could also be used to better evaluate the indications and types of surgical treatment and to plan the postoperative follow-up. AI represents an attractive tool for decision-making and may facilitate development of personalized therapeutic approaches for patients with AAA.

Forecasting the seasonality and trend of pulmonary tuberculosis in Jiangsu Province of China using advanced statistical time-series analyses

Objective

Forecasting the seasonality and trend of pulmonary tuberculosis is important for the rational allocation of health resources; however, this foresting is often hampered by inappropriate prediction methods. In this study, we performed validation research by comparing the accuracy of the autoregressive integrated moving average (ARIMA) model and the back-propagation neural network (BPNN) model in a southeastern province of China.

Methods

We applied the data from 462,214 notified pulmonary tuberculosis cases registered from January 2005 to December 2015 in Jiangsu Province to modulate and construct the ARIMA and BPNN models. Cases registered in 2016 were used to assess the prediction accuracy of the models. The root mean square error (RMSE), mean absolute percentage error (MAPE), mean absolute error (MAE) and mean error rate (MER) were used to evaluate the model fitting and forecasting effect.

Results

During 2005–2015, the annual pulmonary tuberculosis notification rate in Jiangsu Province was 56.35/100,000, ranging from 40.85/100,000 to 79.36/100,000. Through screening and comparison, the ARIMA (0, 1, 2) (0, 1, 1)₁₂ and BPNN (3-9-1) were defined as the optimal fitting models. In the fitting dataset, the RMSE, MAPE, MAE and MER were 0.3901, 6.0498, 0.2740 and 0.0608, respectively, for the ARIMA (0, 1, 2) (0, 1, 1)₁₂ model, 0.3236, 6.0113, 0.2508 and 0.0587, respectively, for the BPNN model. In the forecasting dataset, the RMSE, MAPE, MAE and MER were 0.1758, 4.6041, 0.1368 and 0.0444, respectively, for the ARIMA (0, 1, 2) (0, 1, 1)₁₂ model, and 0.1382, 3.2172, 0.1018 and 0.0330, respectively, for the BPNN model.

Conclusion

Both the ARIMA and BPNN models can be used to predict the seasonality and trend of pulmonary tuberculosis in the Chinese population, but the BPNN model shows better performance. Applying statistical techniques by considering local characteristics may enable more accurate mathematical modeling.

Using multiple classifiers for predicting the risk of endovascular aortic aneurysm repair re-intervention through hybrid feature selection

Feature selection is essential in medical area; however, its process becomes complicated with the presence of censoring which is the unique character of survival analysis. Most survival feature selection methods are based on Cox's proportional hazard model, though machine learning classifiers are preferred. They are less employed in survival analysis due to censoring which prevents them from directly being used to survival data. Among the few work that employed machine learning classifiers, partial logistic artificial neural network with auto-relevance determination is a well-known method that deals with censoring and perform feature selection for survival data. However, it depends on data replication to handle censoring which leads to unbalanced and biased prediction results especially in highly censored data. Other methods cannot deal with high censoring. Therefore, in this article, a new hybrid feature selection method is proposed which presents a solution to high level censoring. It combines support vector machine, neural network, and K-nearest neighbor classifiers using simple majority voting and a new weighted majority voting method based on survival metric to construct a multiple classifier system. The new hybrid feature selection process uses multiple classifier system as a wrapper method and merges it with iterated feature ranking filter method to further reduce features. Two endovascular aortic repair datasets containing 91% censored patients collected from two centers were used to construct a multicenter study to evaluate the performance of the proposed approach. The results showed the proposed technique outperformed individual classifiers and variable selection methods based on Cox's model such as Akaike and Bayesian information criterions and least absolute shrinkage and selector operator in p values of the log-rank test, sensitivity, and concordance index. This indicates that the proposed classifier is more powerful in correctly predicting the risk of re-intervention enabling doctor in selecting patients' future follow-up plan.

Feature selection through validation and un-censoring of endovascular repair survival data for predicting the risk of re-intervention

Background

Feature selection (FS) process is essential in the medical area as it reduces the effort and time needed for physicians to measure unnecessary features. Choosing useful variables is a difficult task with the presence of censoring which is the unique characteristic in survival analysis. Most survival FS methods depend on Cox’s proportional hazard model; however, machine learning techniques (MLT) are preferred but not commonly used due to censoring. Techniques that have been proposed to adopt MLT to perform FS with survival data cannot be used with the high level of censoring. The researcher’s previous publications proposed a technique to deal with the high level of censoring. It also used existing FS techniques to reduce dataset dimension. However, in this paper a new FS technique was proposed and combined with feature transformation and the proposed uncensoring approaches to select a reduced set of features and produce a stable predictive model.

Methods

In this paper, a FS technique based on artificial neural network (ANN) MLT is proposed to deal with highly censored Endovascular Aortic Repair (EVAR). Survival data EVAR datasets were collected during 2004 to 2010 from two vascular centers in order to produce a final stable model. They contain almost 91% of censored patients. The proposed approach used a wrapper FS method with ANN to select a reduced subset of features that predict the risk of EVAR re-intervention after 5 years to patients from two different centers located in the United Kingdom, to allow it to be potentially applied to cross-centers predictions. The proposed model is compared with the two popular FS techniques; Akaike and Bayesian information criteria (AIC, BIC) that are used with Cox’s model.

Results

The final model outperforms other methods in distinguishing the high and low risk groups; as they both have concordance index and estimated AUC better than the Cox’s model based on AIC, BIC, Lasso, and SCAD approaches. These models have p-values lower than 0.05, meaning that patients with different risk groups can be separated significantly and those who would need re-intervention can be correctly predicted.

Conclusion

The proposed approach will save time and effort made by physicians to collect unnecessary variables. The final reduced model was able to predict the long-term risk of aortic complications after EVAR. This predictive model can help clinicians decide patients’ future observation plan.

Electronic supplementary material

The online version of this article (doi:10.1186/s12911-017-0508-3) contains supplementary material, which is available to authorized users.