Artificial neural networks applied to outcome prediction for colorectal cancer patients in separate institutions

To identify important MRI features to differentiate hepatic mucinous cystic neoplasms from septated hepatic cysts based on random forest

OBJECTIVE

To identify important MRI features to differentiate hepatic mucinous cystic neoplasms (MCN) from septated hepatic cysts (HC) using random forest and compared with logistic regression algorithm.

METHODS

Pathologically diagnosed hepatic cysts and hepatic MCNs with pre-operative contrast-enhanced MRI in our hospital from 2010 to 2023 were collected and only septated lesions on enhanced MRI were enrolled. A total of 21 septated HC and 18 MCNs were included in this study. Eighteen MRI features were analyzed and top important features were identified based on random forest (RF) algorithm. The results were evaluated by the prediction performance of a RF model combining the important features and compared with the performance of the logistic regression (LR) algorithm. Finally, for each identified feature, diagnostic probability, sensitivity, and specificity were calculated and compared.

RESULTS

Four variables, i.e., the septation arising from wall without indentation, multiseptate, intracapsular cyst sign, and solitary lesion were extracted as top important features with significance for MCNs by the random forest algorithm. The RF model using these variables had an AUC of 0.982 (0.95CI, 0.950-1.000), compared with the LR model based on two identified features with AUC of 0.931 (0.95CI, 0.846-1.000), p = 0.202. Among the four important features, multiseptate had the highest specificity (95.2%) and good sensitivity (72.2%, lower than the septation from wall without indentation, 94.4%) to diagnose MCNs.

CONCLUSION

Four out of 18 MRI features were extracted as reliably important factors to differ hepatic MCNs from septated HC. The combination of these four features in a RF model could achieve satisfactory diagnostic efficacy.

Evaluation of Matrix Metalloproteases by Artificial Intelligence Techniques in Negative Biopsies as New Diagnostic Strategy in Prostate Cancer

Usually, after an abnormal level of serum prostate-specific antigen (PSA) or digital rectal exam, men undergo a prostate needle biopsy. However, the traditional sextant technique misses 15-46% of cancers. At present, there are problems regarding disease diagnosis/prognosis, especially in patients' classification, because the information to be handled is complex and challenging to process. Matrix metalloproteases (MMPs) have high expression by prostate cancer (PCa) compared with benign prostate tissues. To assess the possible contribution to the diagnosis of PCa, we evaluated the expression of several MMPs in prostate tissues before and after PCa diagnosis using machine learning, classifiers, and supervised algorithms. A retrospective study was conducted on 29 patients diagnosed with PCa with previous benign needle biopsies, 45 patients with benign prostatic hyperplasia (BHP), and 18 patients with high-grade prostatic intraepithelial neoplasia (HGPIN). An immunohistochemical study was performed on tissue samples from tumor and non-tumor areas using specific antibodies against MMP -2, 9, 11, and 13, and the tissue inhibitor of MMPs -3 (TIMP-3), and the protein expression by different cell types was analyzed to which several automatic learning techniques have been applied. Compared with BHP or HGPIN specimens, epithelial cells (ECs) and fibroblasts from benign prostate biopsies before the diagnosis of PCa showed a significantly higher expression of MMPs and TIMP-3. Machine learning techniques provide a differentiable classification between these patients, with greater than 95% accuracy, considering ECs, being slightly lower when considering fibroblasts. In addition, evolutionary changes were found in paired tissues from benign biopsy to prostatectomy specimens in the same patient. Thus, ECs from the tumor zone from prostatectomy showed higher expressions of MMPs and TIMP-3 compared to ECs of the corresponding zone from the benign biopsy. Similar differences were found for expressions of MMP-9 and TIMP-3, between fibroblasts from these zones. The classifiers have determined that patients with benign prostate biopsies before the diagnosis of PCa showed a high MMPs/TIMP-3 expression by ECs, so in the zone without future cancer development as in the zone with future tumor, compared with biopsy samples from patients with BPH or HGPIN. Expression of MMP -2, 9, 11, and 13, and TIMP-3 phenotypically define ECs associated with future tumor development. Also, the results suggest that MMPs/TIMPs expression in biopsy tissues may reflect evolutionary changes from prostate benign tissues to PCa. Thus, these findings in combination with other parameters might contribute to improving the suspicion of PCa diagnosis.

Artificial Neural Network Modeling to Predict Neonatal Metabolic Bone Disease in the Prenatal and Postnatal Periods

Importance

Early recognition of metabolic bone disease (MBD) in infants is necessary but difficult; an appropriate tool to screen infants at risk of developing MBD is needed.

Objectives

To develop a predictive model for neonates at risk for MBD in the prenatal and postnatal periods and detect the pivotal exposed factors in each period.

Design, Setting, and Participants

A diagnostic study was conducted from January 1, 2012, to December 31, 2021, in Shanghai, China. A total of 10 801 pregnant women (singleton pregnancy, followed up until 1 month after parturition) and their infants (n = 10 801) were included. An artificial neural network (ANN) framework was used to build 5 predictive models with different exposures from prenatal to postnatal periods. The receiver operating characteristic curve was used to evaluate the model performance. The importance of each feature was examined and ranked.

Results

Of the 10 801 Chinese women who participated in the study (mean [SD] age, 29.7 [3.9] years), 7104 (65.8%) were local residents, 1001 (9.3%) had uterine scarring, and 138 (1.3%) gave birth to an infant with MBD. Among the 5 ANN models, model 1 (significant prenatal and postnatal factors) showed the highest AUC of 0.981 (95% CI, 0.970-0.992), followed by model 5 (postnatal factors; AUC, 0.977; 95% CI, 0.966-0.988), model 4 (all prenatal factors; AUC, 0.850; 95% CI, 0.785-0.915), model 3 (gestational complications or comorbidities and medication use; AUC, 0.808; 95% CI, 0.726-0.891), and model 2 (maternal nutritional conditions; AUC, 0.647; 95% CI, 0.571-0.723). Birth weight, maternal age at pregnancy, and neonatal disorders (anemia, respiratory distress syndrome, and septicemia) were the most important model 1 characteristics for predicting infants at risk of MBD; among these characteristics, extremely low birth weight (importance, 50.5%) was the most powerful factor. The use of magnesium sulfate during pregnancy (model 4: importance, 21.2%) was the most significant predictor of MBD risk in the prenatal period.

Conclusions and Relevance

In this diagnostic study, ANN appeared to be a simple and efficient tool for identifying neonates at risk for MBD. Combining prenatal and postnatal factors or using postnatal exposures alone provided the most precise prediction. Extremely low birth weight was the most significant predictive factor, whereas magnesium sulfate use during pregnancy could be an important bellwether for MBD before delivery.

Comprehensive Review on the Use of Artificial Intelligence in Ophthalmology and Future Research Directions

BACKGROUND

Having several applications in medicine, and in ophthalmology in particular, artificial intelligence (AI) tools have been used to detect visual function deficits, thus playing a key role in diagnosing eye diseases and in predicting the evolution of these common and disabling diseases. AI tools, i.e., artificial neural networks (ANNs), are progressively involved in detecting and customized control of ophthalmic diseases. The studies that refer to the efficiency of AI in medicine and especially in ophthalmology were analyzed in this review.

MATERIALS AND METHODS

We conducted a comprehensive review in order to collect all accounts published between 2015 and 2022 that refer to these applications of AI in medicine and especially in ophthalmology. Neural networks have a major role in establishing the demand to initiate preliminary anti-glaucoma therapy to stop the advance of the disease.

RESULTS

Different surveys in the literature review show the remarkable benefit of these AI tools in ophthalmology in evaluating the visual field, optic nerve, and retinal nerve fiber layer, thus ensuring a higher precision in detecting advances in glaucoma and retinal shifts in diabetes. We thus identified 1762 applications of artificial intelligence in ophthalmology: review articles and research articles (301 pub med, 144 scopus, 445 web of science, 872 science direct). Of these, we analyzed 70 articles and review papers (diabetic retinopathy (N = 24), glaucoma (N = 24), DMLV (N = 15), other pathologies (N = 7)) after applying the inclusion and exclusion criteria.

CONCLUSION

In medicine, AI tools are used in surgery, radiology, gynecology, oncology, etc., in making a diagnosis, predicting the evolution of a disease, and assessing the prognosis in patients with oncological pathologies. In ophthalmology, AI potentially increases the patient's access to screening/clinical diagnosis and decreases healthcare costs, mainly when there is a high risk of disease or communities face financial shortages. AI/DL (deep learning) algorithms using both OCT and FO images will change image analysis techniques and methodologies. Optimizing these (combined) technologies will accelerate progress in this area.

A Machine Learning Approach Using XGBoost Predicts Lung Metastasis in Patients with Ovarian Cancer

Background

Liver metastasis (LM) is an independent risk factor that affects the prognosis of patients with ovarian cancer; however, there is still a lack of prediction. This study developed a limit gradient enhancement (XGBoost) to predict the risk of lung metastasis in newly diagnosed patients with ovarian cancer, thereby improving prediction efficiency. Patients and Methods. Data of patients diagnosed with ovarian cancer in the Surveillance, Epidemiology, and Final Results (SEER) database from 2010 to 2015 were retrospectively collected. The XGBoost algorithm was used to establish a lung metastasis model for patients with ovarian cancer. The performance of the predictive model was tested by the area under the curve (AUC) of the receiver operating characteristic curve (ROC).

Results

The results of the XGBoost algorithm showed that the top five important factors were age, laterality, histological type, grade, and marital status. XGBoost showed good discriminative ability, with an AUC of 0.843. Accuracy, sensitivity, and specificity were 0.982, 1.000, and 0.686, respectively.

Conclusion

This study is the first to develop a machine-learning-based prediction model for lung metastasis in patients with ovarian cancer. The prediction model based on the XGBoost algorithm has a higher accuracy rate than traditional logistic regression and can be used to predict the risk of lung metastasis in newly diagnosed patients with ovarian cancer.

Development and Validation of a Deep Learning Model for Predicting Treatment Response in Patients With Newly Diagnosed Epilepsy

Importance

Selection of antiseizure medications (ASMs) for epilepsy remains largely a trial-and-error approach. Under this approach, many patients have to endure sequential trials of ineffective treatments until the "right drugs" are prescribed.

Objective

To develop and validate a deep learning model using readily available clinical information to predict treatment success with the first ASM for individual patients.

Design, Setting, and Participants

This cohort study developed and validated a prognostic model. Patients were treated between 1982 and 2020. All patients were followed up for a minimum of 1 year or until failure of the first ASM. A total of 2404 adults with epilepsy newly treated at specialist clinics in Scotland, Malaysia, Australia, and China between 1982 and 2020 were considered for inclusion, of whom 606 (25.2%) were excluded from the final cohort because of missing information in 1 or more variables.

Exposures

One of 7 antiseizure medications.

Main Outcomes and Measures

With the use of the transformer model architecture on 16 clinical factors and ASM information, this cohort study first pooled all cohorts for model training and testing. The model was trained again using the largest cohort and externally validated on the other 4 cohorts. The area under the receiver operating characteristic curve (AUROC), weighted balanced accuracy, sensitivity, and specificity of the model were all assessed for predicting treatment success based on the optimal probability cutoff. Treatment success was defined as complete seizure freedom for the first year of treatment while taking the first ASM. Performance of the transformer model was compared with other machine learning models.

Results

The final pooled cohort included 1798 adults (54.5% female; median age, 34 years [IQR, 24-50 years]). The transformer model that was trained using the pooled cohort had an AUROC of 0.65 (95% CI, 0.63-0.67) and a weighted balanced accuracy of 0.62 (95% CI, 0.60-0.64) on the test set. The model that was trained using the largest cohort only had AUROCs ranging from 0.52 to 0.60 and a weighted balanced accuracy ranging from 0.51 to 0.62 in the external validation cohorts. Number of pretreatment seizures, presence of psychiatric disorders, electroencephalography, and brain imaging findings were the most important clinical variables for predicted outcomes in both models. The transformer model that was developed using the pooled cohort outperformed 2 of the 5 other models tested in terms of AUROC.

Conclusions and Relevance

In this cohort study, a deep learning model showed the feasibility of personalized prediction of response to ASMs based on clinical information. With improvement of performance, such as by incorporating genetic and imaging data, this model may potentially assist clinicians in selecting the right drug at the first trial.

Using Machine Learning to Predict Complications in Pregnancy: A Systematic Review

Introduction: Artificial intelligence is widely used in medical field, and machine learning has been increasingly used in health care, prediction, and diagnosis and as a method of determining priority. Machine learning methods have been features of several tools in the fields of obstetrics and childcare. This present review aims to summarize the machine learning techniques to predict perinatal complications.

Objective: To identify the applicability and performance of machine learning methods used to identify pregnancy complications.

Methods: A total of 98 articles were obtained with the keywords “machine learning,” “deep learning,” “artificial intelligence,” and accordingly as they related to perinatal complications (“complications in pregnancy,” “pregnancy complications”) from three scientific databases: PubMed, Scopus, and Web of Science. These were managed on the Mendeley platform and classified using the PRISMA method.

Results: A total of 31 articles were selected after elimination according to inclusion and exclusion criteria. The features used to predict perinatal complications were primarily electronic medical records (48%), medical images (29%), and biological markers (19%), while 4% were based on other types of features, such as sensors and fetal heart rate. The main perinatal complications considered in the application of machine learning thus far are pre-eclampsia and prematurity. In the 31 studies, a total of sixteen complications were predicted. The main precision metric used is the AUC. The machine learning methods with the best results were the prediction of prematurity from medical images using the support vector machine technique, with an accuracy of 95.7%, and the prediction of neonatal mortality with the XGBoost technique, with 99.7% accuracy.

Conclusion: It is important to continue promoting this area of research and promote solutions with multicenter clinical applicability through machine learning to reduce perinatal complications. This systematic review contributes significantly to the specialized literature on artificial intelligence and women’s health.

Deep Learning Neural Modelling as a Precise Method in the Assessment of the Chronological Age of Children and Adolescents Using Tooth and Bone Parameters

Dental age is one of the most reliable methods for determining a patient’s age. The timing of teething, the period of tooth replacement, or the degree of tooth attrition is an important diagnostic factor in the assessment of an individual’s developmental age. It is used in orthodontics, pediatric dentistry, endocrinology, forensic medicine, and pathomorphology, but also in scenarios regarding international adoptions and illegal immigrants. The methods used to date are time-consuming and not very precise. For this reason, artificial intelligence methods are increasingly used to estimate the age of a patient. The present work is a continuation of the work of Zaborowicz et al. In the presented research, a set of 21 original indicators was used to create deep neural network models. The aim of this study was to verify the ability to generate a more accurate deep neural network model compared to models produced previously. The quality parameters of the produced models were as follows. The MAE error of the produced models, depending on the learning set used, was between 2.34 and 4.61 months, while the RMSE error was between 5.58 and 7.49 months. The correlation coefficient R² ranged from 0.92 to 0.96.

Deep Machine Learning for Oral Cancer: From Precise Diagnosis to Precision Medicine

Oral squamous cell carcinoma (OSCC) is one of the most prevalent cancers worldwide and its incidence is on the rise in many populations. The high incidence rate, late diagnosis, and improper treatment planning still form a significant concern. Diagnosis at an early-stage is important for better prognosis, treatment, and survival. Despite the recent improvement in the understanding of the molecular mechanisms, late diagnosis and approach toward precision medicine for OSCC patients remain a challenge. To enhance precision medicine, deep machine learning technique has been touted to enhance early detection, and consequently to reduce cancer-specific mortality and morbidity. This technique has been reported to have made a significant progress in data extraction and analysis of vital information in medical imaging in recent years. Therefore, it has the potential to assist in the early-stage detection of oral squamous cell carcinoma. Furthermore, automated image analysis can assist pathologists and clinicians to make an informed decision regarding cancer patients. This article discusses the technical knowledge and algorithms of deep learning for OSCC. It examines the application of deep learning technology in cancer detection, image classification, segmentation and synthesis, and treatment planning. Finally, we discuss how this technique can assist in precision medicine and the future perspective of deep learning technology in oral squamous cell carcinoma.

Artificial Intelligence and Machine Learning in Cancer Research: A Systematic and Thematic Analysis of the Top 100 Cited Articles Indexed in Scopus Database

INTRODUCTION

Cancer is a major public health problem and a global leading cause of death where the screening, diagnosis, prediction, survival estimation, and treatment of cancer and control measures are still a major challenge. The rise of Artificial Intelligence (AI) and Machine Learning (ML) techniques and their applications in various fields have brought immense value in providing insights into advancement in support of cancer control.

METHODS

A systematic and thematic analysis was performed on the Scopus database to identify the top 100 cited articles in cancer research. Data were analyzed using RStudio and VOSviewer.Var1.6.6.

RESULTS

The top 100 articles in AI and ML in cancer received a 33 920 citation score with a range of 108 to 5758 times. Doi Kunio from the USA was the most cited author with total number of citations (TNC = 663). Out of 43 contributed countries, 30% of the top 100 cited articles originated from the USA, and 10% originated from China. Among the 57 peer-reviewed journals, the "Expert Systems with Application" published 8% of the total articles. The results were presented in highlight technological advancement through AI and ML via the widespread use of Artificial Neural Network (ANNs), Deep Learning or machine learning techniques, Mammography-based Model, Convolutional Neural Networks (SC-CNN), and text mining techniques in the prediction, diagnosis, and prevention of various types of cancers towards cancer control.

CONCLUSIONS

This bibliometric study provides detailed overview of the most cited empirical evidence in AI and ML adoption in cancer research that could efficiently help in designing future research. The innovations guarantee greater speed by using AI and ML in the detection and control of cancer to improve patient experience.

Artificial neural network for the prediction model of glomerular filtration rate to estimate the normal or abnormal stages of kidney using gamma camera

OBJECTIVE

Chronic kidney disease (CKD) is evaluated based on glomerular filtration rate (GFR) using a gamma camera in the nuclear medicine center or hospital in a routine procedure, but the gamma camera does not provide the accurate stages of the diseases. Therefore, this research aimed to find out the normal or abnormal stages of CKD based on the value of GFR using an artificial neural network (ANN).

METHODS

Two hundred fifty (Training 188, Testing 62) kidney patients who underwent the ultrasonography test to diagnose the renal test in our nuclear medical centre were scanned using gamma camera. The patients were injected with 99mTc-DTPA before the scanning procedure. After pushing the syringe into the patient's vein, the pre-syringe and post syringe radioactive counts were calculated using the gamma camera. The artificial neural network uses the softmax function with cross-entropy loss to diagnose CKD normal or abnormal labels depending on the value of GFR in the output layer.

RESULTS

The results showed that the accuracy of the proposed ANN model was 99.20% for K-fold cross-validation. The sensitivity and specificity were 99.10% and 99.20%, respectively. The Area under the curve (AUC) was 0.9994.

CONCLUSION

The proposed model using an artificial neural network can classify the normal or abnormal stages of CKD. After implementing the proposed model clinically, it may upgrade the gamma camera to diagnose the normal or abnormal stages of the CKD with an appropriate GFR value.

BH-index: A predictive system based on serum biomarkers and ensemble learning for early colorectal cancer diagnosis in mass screening

BACKGROUND AND OBJECTIVE

Colorectal cancer is one of the most common malignancies among the general population. Artificial Intelligence methodologies based on serum parameters are in continuous development to obtain less expensive tools for highly sensitive diagnoses. This study proposes a predictive system based on serum biomarkers and ensemble learning to predict colorectal cancer presence and the related TNM stage in patients.

METHODS

We have selected 17 significant plasmatic proteins, i.e., Carcinoembryonic Antigen, CA 19-9, CA 125, CA 50, CA 72-4, Tissue Polypeptide Antigen, C-Reactive Protein, Ceruloplasmin, Haptoglobin, Transferrin, Ferritin, α-1-Antitrypsin, α-2-Macroglobulin, α-1 Acid Glycoprotein, Complement C4, Complement C3, and Retinol Binding Protein, regarding 345 patients (248 affected by the neoplastic disease). The proposed system consists of two predictors, i.e., binary and staging; the former predicts the presence/absence of cancer, while the latter identifies the related TNM stage (I, II, III, or IV). The experiments were conducted by deploying and comparing Random Forest, XGBoost, Support Vector Machine, and Multilayer Perceptron with feature selection based on Gini Importance and with dimensionality reduction via PCA.

RESULTS

The results show that the system composed of XGBoost as binary and staging predictor reaches 91.30% accuracy, 90% sensitivity, and 93.33% specificity for the absence/presence outcome, while 66.66% accuracy for the staging response. With the expansion of the training set in favor of positive patients and majority voting, the system composed of the combination of Support Vector Machine, XGBoost, and Multilayer Perceptron as the binary predictor reaches 98.03% accuracy, 100% sensitivity, and 92.30% specificity, while the combination of Random Forest, XGBoost, and Multilayer Perceptron as staging predictor achieves 60% accuracy. The final system reaches, in terms of accuracy, 98.03%, and 66.66% for the binary and staging predictors, respectively. It was also found that the biomarkers which contribute most to the binary decision are Ceruloplasmin and α-2-Macroglobulin, while the least significant dimensions are CA 50 and α-1-Antitrypsin; instead, Carcinoembryonic Antigen and α-1 Acid Glycoprotein are the most significant to the staging decision.

CONCLUSIONS

The present study proves the effectiveness of deploying serum biomarkers as feature dimensions for early colorectal cancer diagnosis and of using majority voting for noise reduction in the prediction.

Tooth and Bone Parameters in the Assessment of the Chronological Age of Children and Adolescents Using Neural Modelling Methods

The analog methods used in the clinical assessment of the patient's chronological age are subjective and characterized by low accuracy. When using those methods, there is a noticeable discrepancy between the chronological age and the age estimated based on relevant scientific studies. Innovations in the field of information technology are increasingly used in medicine, with particular emphasis on artificial intelligence methods. The paper presents research aimed at developing a new, effective methodology for the assessment of the chronological age using modern IT methods. In this paper, a study was conducted to determine the features of pantomographic images that support the determination of metric age, and neural models were produced to support the process of identifying the age of children and adolescents. The whole conducted work was a new methodology of metric age assessment. The result of the conducted study is a set of 21 original indicators necessary for the assessment of the chronological age with the use of computer image analysis and neural modelling, as well as three non-linear models of radial basis function networks (RBF), whose accuracy ranges from 96 to 99%. The result of the research are three neural models that determine the chronological age.

Risk Stratification Using 18F-FDG PET/CT and Artificial Neural Networks in Head and Neck Cancer Patients Undergoing Radiotherapy

This study retrospectively analyzed the performance of artificial neural networks (ANN) to predict overall survival (OS) or locoregional failure (LRF) in HNSCC patients undergoing radiotherapy, based on 2-[¹⁸F]FDG PET/CT and clinical covariates. We compared predictions relying on three different sets of features, extracted from 230 patients. Specifically, (i) an automated feature selection method independent of expert rating was compared with (ii) clinical variables with proven influence on OS or LRF and (iii) clinical data plus expert-selected SUV metrics. The three sets were given as input to an artificial neural network for outcome prediction, evaluated by Harrell’s concordance index (HCI) and by testing stratification capability. For OS and LRF, the best performance was achieved with expert-based PET-features (0.71 HCI) and clinical variables (0.70 HCI), respectively. For OS stratification, all three feature sets were significant, whereas for LRF only expert-based PET-features successfully classified low vs. high-risk patients. Based on 2-[¹⁸F]FDG PET/CT features, stratification into risk groups using ANN for OS and LRF is possible. Differences in the results for different feature sets confirm the relevance of feature selection, and the key importance of expert knowledge vs. automated selection.

Radiomics AI prediction for head and neck squamous cell carcinoma (HNSCC) prognosis and recurrence with target volume approach

OBJECTIVES

To evaluate the performance of radiomics features extracted from planning target volume (PTV) and gross tumor volume (GTV) in the prediction of the death prognosis and cancer recurrence rate for head and neck squamous cell carcinoma (HNSCC).

METHODS

188 HNSCC patients' planning CT images with radiotherapy structures sets were acquired from Cancer Imaging Archive (TCIA). The 3D slicer (v. 4.10.2) with the PyRadiomics extension (Computational Imaging and Bioinformatics Lab, Harvard medical School) was used to extract radiomics features from the radiotherapy planning images. An in-house developed deep learning artificial neural networks (DL-ANN) model was used to predict death prognosis and cancer recurrence rate based on the features extracted from GTV and PTV of the CT images.

RESULTS

The PTV radiomics features with DL-ANN model could achieve 77.7% accuracy with overall AUC equal to 0.934 and 0.932 when predicting HNSCC-related death prognosis and cancer recurrence respectively. Furthermore, the DL-ANN model can achieve an accuracy of 74.3% with AUC equal to 0.947 and 0.956 for the HNSCC-related death prognosis and cancer recurrence respectively using GTV features.

CONCLUSION

Using both GTV and PTV radiomics features in the DL-ANN model, can aid in predicting HNSCC-related death prognosis and cancer recurrence. Clinicians may find it helpful in formulating different treatment regimens and facilitate personized medicine based on the predicted outcome when performing GTV and PTV delineation.

ADVANCES IN KNOWLEDGE

Radiomics features of GTV and PTV are reliable prognosis and recurrence predicting tools, which may help clinicians in GTV and PTV delineation to facilitate delivery of personalized treatment.

Status quo and future prospects of artificial neural network from the perspective of gastroenterologists

Artificial neural networks (ANNs) are one of the primary types of artificial intelligence and have been rapidly developed and used in many fields. In recent years, there has been a sharp increase in research concerning ANNs in gastrointestinal (GI) diseases. This state-of-the-art technique exhibits excellent performance in diagnosis, prognostic prediction, and treatment. Competitions between ANNs and GI experts suggest that efficiency and accuracy might be compatible in virtue of technique advancements. However, the shortcomings of ANNs are not negligible and may induce alterations in many aspects of medical practice. In this review, we introduce basic knowledge about ANNs and summarize the current achievements of ANNs in GI diseases from the perspective of gastroenterologists. Existing limitations and future directions are also proposed to optimize ANN’s clinical potential. In consideration of barriers to interdisciplinary knowledge, sophisticated concepts are discussed using plain words and metaphors to make this review more easily understood by medical practitioners and the general public.

Real-time phase-retrieval and wavefront sensing enabled by an artificial neural network

In this manuscript we demonstrate a method to reconstruct the wavefront of focused beams from a measured diffraction pattern behind a diffracting mask in real-time. The phase problem is solved by means of a neural network, which is trained with simulated data and verified with experimental data. The neural network allows live reconstructions within a few milliseconds, which previously with iterative phase retrieval took several seconds, thus allowing the adjustment of complex systems and correction by adaptive optics in real time. The neural network additionally outperforms iterative phase retrieval with high noise diffraction patterns.

Machine learning for the prediction of bone metastasis in patients with newly diagnosed thyroid cancer

OBJECTIVES

This study aimed to establish a machine learning prediction model that can be used to predict bone metastasis (BM) in patients with newly diagnosed thyroid cancer (TC).

METHODS

Demographic and clinicopathologic variables of TC patients in the Surveillance, Epidemiology, and End Results database from 2010 to 2016 were retrospectively analyzed. On this basis, we developed a random forest (RF) algorithm model based on machine-learning. The area under receiver operating characteristic curve (AUC), accuracy score, recall rate, and specificity are used to evaluate and compare the prediction performance of the RF model and the other model.

RESULTS

A total of 17,138 patients were included in the study, with 166 (0.97%) developed bone metastases. Grade, T stage, histology, race, sex, age, and N stage were the important prediction features of BM. The RF model has better predictive performance than the other model (AUC: 0.917, accuracy: 0.904, recall rate: 0.833, and specificity: 0.905).

CONCLUSIONS

The RF model constructed in this study could accurately predict bone metastases in TC patients, which may provide clinicians with more personalized clinical decision-making recommendations. Machine learning technology has the potential to improve the development of BM prediction models in TC patients.

Application of artificial intelligence and machine learning for prediction of oral cancer risk

BACKGROUND

Oral cancer requires early diagnosis and treatment to increase the chances of survival. This study aimed to develop an artificial neural network model that helps to predict the individuals' risk of developing oral cancer based on data on risk factors, systematic medical condition, and clinic-pathological features.

METHODS

A popular data mining algorithm artificial neural network was used for developing the artificial intelligence-based prediction model. A total of 29 variables that were associated with the patients were used for developing the model. The dataset was randomly split into the training dataset 54 (75%) cases and testing dataset 19 (25%) cases. All records and observations were reviewed by Board-certified oral pathologist.

RESULTS

A total of 73 patients met the eligibility criteria. Twenty-two (30.13%) were benign cases, and 51 (69.86%) were malignant cases. Thirty-seven were female, and 36 were male, with a mean age of 63.09 years. Our analysis displayed that the average sensitivity and specificity of ANN for oral cancer prediction based on the 10-fold cross-validation analysis was 85.71% (95% confidence interval [CI], 57.19-98.22) and 60.00% (95% CI, 14.66-94.73), respectively. The accuracy of ANN for oral cancer prediction was 78.95% (95% CI, 54.43-931.95).

CONCLUSION

Our results suggest that this machine-learning technique has the potential to help in oral cancer screening and diagnosis based on the datasets. The results demonstrate that the artificial neural network could perform well in estimating the probability of malignancy and improve the positive predictive value that could help to predict the individuals' risk of developing OC based on knowledge of their risk factors, systemic medical conditions, and clinic-pathological data.

Relative importance of clinical and sociodemographic factors in association with post-operative in-hospital deaths in colorectal cancer patients in New South Wales: An artificial neural network approach

RATIONALE, AIMS AND OBJECTIVES

Co-morbidities in colorectal cancer patients complicate hospital care, and their relative importance to post-operative deaths is largely unknown. This study was conducted to examine a range of clinical and sociodemographic factors in relation to post-operative in-hospital deaths in colorectal cancer patients and identify whether these contributions would vary by severity of co-morbidities.

METHODS

In this multicentre retrospective cohort study, we used the complete census of New South Wales inpatient data to select colorectal cancer patients admitted to public hospitals for acute surgical care, who underwent procedures on the digestive system during the period of July 2001 to June 2014. The primary outcome was in-hospital death at the end of acute care. Multilayer perceptron and back-propagation artificial neural networks (ANNs) were used to quantify the relative importance of a wide range of clinical and sociodemographic factors in relation to post-operative deaths, stratified by severity of co-morbidities based on Charlson co-morbidity index.

RESULTS

Of 6288 colorectal cancer patients, approximately 58.3% (n = 3669) had moderate to severe co-morbidities. A total of 464 (7.4%) died in hospitals. The performance for ANN models was superior to logistic models. Co-morbid musculoskeletal and mental disorders, adverse events in health care, and socio-economic factors including rural residence and private insurance status contributed to post-operative deaths in hospitals.

CONCLUSION

Identification of relative importance of factors contributing to in-hospital deaths in colorectal cancer patients using ANN may help to enhance patient-centred strategies to meet complex needs during acute surgical care and prevent post-operative in-hospital deaths.

Machine learning model to predict oncologic outcomes for drugs in randomized clinical trials

Predicting oncologic outcome is challenging due to the diversity of cancer histologies and the complex network of underlying biological factors. In this study, we determine whether machine learning (ML) can extract meaningful associations between oncologic outcome and clinical trial, drug-related biomarker and molecular profile information. We analyzed therapeutic clinical trials corresponding to 1102 oncologic outcomes from 104 758 cancer patients with advanced colorectal adenocarcinoma, pancreatic adenocarcinoma, melanoma and nonsmall-cell lung cancer. For each intervention arm, a dataset with the following attributes was curated: line of treatment, the number of cytotoxic chemotherapies, small-molecule inhibitors, or monoclonal antibody agents, drug class, molecular alteration status of the clinical arm's population, cancer type, probability of drug sensitivity (PDS) (integrating the status of genomic, transcriptomic and proteomic biomarkers in the population of interest) and outcome. A total of 467 progression-free survival (PFS) and 369 overall survival (OS) data points were used as training sets to build our ML (random forest) model. Cross-validation sets were used for PFS and OS, obtaining correlation coefficients (r) of 0.82 and 0.70, respectively (outcome vs model's parameters). A total of 156 PFS and 110 OS data points were used as test sets. The Spearman correlation (r_s ) between predicted and actual outcomes was statistically significant (PFS: r_s = 0.879, OS: r_s = 0.878, P < .0001). The better outcome arm was predicted in 81% (PFS: N = 59/73, z = 5.24, P < .0001) and 71% (OS: N = 37/52, z = 2.91, P = .004) of randomized trials. The success of our algorithm to predict clinical outcome may be exploitable as a model to optimize clinical trial design with pharmaceutical agents.

Artificial Intelligence to Decode Cancer Mechanism: Beyond Patient Stratification for Precision Oncology

The multitude of multi-omics data generated cost-effectively using advanced high-throughput technologies has imposed challenging domain for research in Artificial Intelligence (AI). Data curation poses a significant challenge as different parameters, instruments, and sample preparations approaches are employed for generating these big data sets. AI could reduce the fuzziness and randomness in data handling and build a platform for the data ecosystem, and thus serve as the primary choice for data mining and big data analysis to make informed decisions. However, AI implication remains intricate for researchers/clinicians lacking specific training in computational tools and informatics. Cancer is a major cause of death worldwide, accounting for an estimated 9.6 million deaths in 2018. Certain cancers, such as pancreatic and gastric cancers, are detected only after they have reached their advanced stages with frequent relapses. Cancer is one of the most complex diseases affecting a range of organs with diverse disease progression mechanisms and the effectors ranging from gene-epigenetics to a wide array of metabolites. Hence a comprehensive study, including genomics, epi-genomics, transcriptomics, proteomics, and metabolomics, along with the medical/mass-spectrometry imaging, patient clinical history, treatments provided, genetics, and disease endemicity, is essential. Cancer Moonshot℠ Research Initiatives by NIH National Cancer Institute aims to collect as much information as possible from different regions of the world and make a cancer data repository. AI could play an immense role in (a) analysis of complex and heterogeneous data sets (multi-omics and/or inter-omics), (b) data integration to provide a holistic disease molecular mechanism, (c) identification of diagnostic and prognostic markers, and (d) monitor patient's response to drugs/treatments and recovery. AI enables precision disease management well beyond the prevalent disease stratification patterns, such as differential expression and supervised classification. This review highlights critical advances and challenges in omics data analysis, dealing with data variability from lab-to-lab, and data integration. We also describe methods used in data mining and AI methods to obtain robust results for precision medicine from "big" data. In the future, AI could be expanded to achieve ground-breaking progress in disease management.

Artificial intelligence in cancer diagnosis and prognosis: Opportunities and challenges

Cancer is an aggressive disease with a low median survival rate. Ironically, the treatment process is long and very costly due to its high recurrence and mortality rates. Accurate early diagnosis and prognosis prediction of cancer are essential to enhance the patient's survival rate. Developments in statistics and computer engineering over the years have encouraged many scientists to apply computational methods such as multivariate statistical analysis to analyze the prognosis of the disease, and the accuracy of such analyses is significantly higher than that of empirical predictions. Furthermore, as artificial intelligence (AI), especially machine learning and deep learning, has found popular applications in clinical cancer research in recent years, cancer prediction performance has reached new heights. This article reviews the literature on the application of AI to cancer diagnosis and prognosis, and summarizes its advantages. We explore how AI assists cancer diagnosis and prognosis, specifically with regard to its unprecedented accuracy, which is even higher than that of general statistical applications in oncology. We also demonstrate ways in which these methods are advancing the field. Finally, opportunities and challenges in the clinical implementation of AI are discussed. Hence, this article provides a new perspective on how AI technology can help improve cancer diagnosis and prognosis, and continue improving human health in the future.

Prediction model development of late-onset preeclampsia using machine learning-based methods

Preeclampsia is one of the leading causes of maternal and fetal morbidity and mortality. Due to the lack of effective preventive measures, its prediction is essential to its prompt management. This study aimed to develop models using machine learning to predict late-onset preeclampsia using hospital electronic medical record data. The performance of the machine learning based models and models using conventional statistical methods were also compared. A total of 11,006 pregnant women who received antenatal care at Yonsei University Hospital were included. Maternal data were retrieved from electronic medical records during the early second trimester to 34 weeks. The prediction outcome was late-onset preeclampsia occurrence after 34 weeks’ gestation. Pattern recognition and cluster analysis were used to select the parameters included in the prediction models. Logistic regression, decision tree model, naïve Bayes classification, support vector machine, random forest algorithm, and stochastic gradient boosting method were used to construct the prediction models. C-statistics was used to assess the performance of each model. The overall preeclampsia development rate was 4.7% (474 patients). Systolic blood pressure, serum blood urea nitrogen and creatinine levels, platelet counts, serum potassium level, white blood cell count, serum calcium level, and urinary protein were the most influential variables included in the prediction models. C-statistics for the decision tree model, naïve Bayes classification, support vector machine, random forest algorithm, stochastic gradient boosting method, and logistic regression models were 0.857, 0.776, 0.573, 0.894, 0.924, and 0.806, respectively. The stochastic gradient boosting model had the best prediction performance with an accuracy and false positive rate of 0.973 and 0.009, respectively. The combined use of maternal factors and common antenatal laboratory data of the early second trimester through early third trimester could effectively predict late-onset preeclampsia using machine learning algorithms. Future prospective studies are needed to verify the clinical applicability algorithms.

Machine learning and its potential applications to the genomic study of head and neck cancer-A systematic review

BACKGROUND

Machine learning (ML) is powerful tool that can identify and classify patterns from large quantities of cancer genomic data that may lead to the discovery of new biomarkers, new drug targets, and a better understanding of important cancer genes. The aim of this systematic review was to evaluate the existing literature and assess the application of machine learning of genomic data in head and neck cancer (HNC).

MATERIALS AND METHODS

The addressed focused question was "Does machine learning of genomic data play a role in prognostic prediction of HNC?" PubMed, EMBASE, Scopus, Web of Science, and gray literature from January 1990 up to and including May 2018 were searched. Two independent reviewers performed the study selection according to eligibility criteria.

RESULTS

A total of seven studies that met the eligibility criteria were included. The majority of studies were cohort studies, one a case-control study and one a randomized controlled trial. Two studies each evaluated oral cancer and laryngeal cancer, while other one study each evaluated nasopharyngeal cancer and oropharyngeal cancer. The majority of studies employed support vector machine (SVM) as a ML technique. Among the included studies, the accuracy rates for ML techniques ranged from 56.7% to 99.4%.

CONCLUSION

Our findings showed that ML techniques for the analysis of genomic data can play a role in the prognostic prediction of HNC.

Deep learning based tissue analysis predicts outcome in colorectal cancer

Image-based machine learning and deep learning in particular has recently shown expert-level accuracy in medical image classification. In this study, we combine convolutional and recurrent architectures to train a deep network to predict colorectal cancer outcome based on images of tumour tissue samples. The novelty of our approach is that we directly predict patient outcome, without any intermediate tissue classification. We evaluate a set of digitized haematoxylin-eosin-stained tumour tissue microarray (TMA) samples from 420 colorectal cancer patients with clinicopathological and outcome data available. The results show that deep learning-based outcome prediction with only small tissue areas as input outperforms (hazard ratio 2.3; CI 95% 1.79-3.03; AUC 0.69) visual histological assessment performed by human experts on both TMA spot (HR 1.67; CI 95% 1.28-2.19; AUC 0.58) and whole-slide level (HR 1.65; CI 95% 1.30-2.15; AUC 0.57) in the stratification into low- and high-risk patients. Our results suggest that state-of-the-art deep learning techniques can extract more prognostic information from the tissue morphology of colorectal cancer than an experienced human observer.

Prediction of 5-year overall survival in cervical cancer patients treated with radical hysterectomy using computational intelligence methods

BACKGROUND

Computational intelligence methods, including non-linear classification algorithms, can be used in medical research and practice as a decision making tool. This study aimed to evaluate the usefulness of artificial intelligence models for 5-year overall survival prediction in patients with cervical cancer treated by radical hysterectomy.

METHODS

The data set was collected from 102 patients with cervical cancer FIGO stage IA2-IIB, that underwent primary surgical treatment. Twenty-three demographic, tumor-related parameters and selected perioperative data of each patient were collected. The simulations involved six computational intelligence methods: the probabilistic neural network (PNN), multilayer perceptron network, gene expression programming classifier, support vector machines algorithm, radial basis function neural network and k-Means algorithm. The prediction ability of the models was determined based on the accuracy, sensitivity, specificity, as well as the area under the receiver operating characteristic curve. The results of the computational intelligence methods were compared with the results of linear regression analysis as a reference model.

RESULTS

The best results were obtained by the PNN model. This neural network provided very high prediction ability with an accuracy of 0.892 and sensitivity of 0.975. The area under the receiver operating characteristics curve of PNN was also high, 0.818. The outcomes obtained by other classifiers were markedly worse.

CONCLUSIONS

The PNN model is an effective tool for predicting 5-year overall survival in cervical cancer patients treated with radical hysterectomy.

Applications of Machine Learning in Cancer Prediction and Prognosis

Machine learning is a branch of artificial intelligence that employs a variety of statistical, probabilistic and optimization techniques that allows computers to “learn” from past examples and to detect hard-to-discern patterns from large, noisy or complex data sets. This capability is particularly well-suited to medical applications, especially those that depend on complex proteomic and genomic measurements. As a result, machine learning is frequently used in cancer diagnosis and detection. More recently machine learning has been applied to cancer prognosis and prediction. This latter approach is particularly interesting as it is part of a growing trend towards personalized, predictive medicine. In assembling this review we conducted a broad survey of the different types of machine learning methods being used, the types of data being integrated and the performance of these methods in cancer prediction and prognosis. A number of trends are noted, including a growing dependence on protein biomarkers and microarray data, a strong bias towards applications in prostate and breast cancer, and a heavy reliance on “older” technologies such artificial neural networks (ANNs) instead of more recently developed or more easily interpretable machine learning methods. A number of published studies also appear to lack an appropriate level of validation or testing. Among the better designed and validated studies it is clear that machine learning methods can be used to substantially (15–25%) improve the accuracy of predicting cancer susceptibility, recurrence and mortality. At a more fundamental level, it is also evident that machine learning is also helping to improve our basic understanding of cancer development and progression.

Microdevice Platform for In Vitro Nervous System and Its Disease Model

The development of precise microdevices can be applied to the reconstruction of in vitro human microenvironmental systems with biomimetic physiological conditions that have highly tunable spatial and temporal features. Organ-on-a-chip can emulate human physiological functions, particularly at the organ level, as well as its specific roles in the body. Due to the complexity of the structure of the central nervous system and its intercellular interaction, there remains an urgent need for the development of human brain or nervous system models. Thus, various microdevice models have been proposed to mimic actual human brain physiology, which can be categorized as nervous system-on-a-chip. Nervous system-on-a-chip platforms can prove to be promising technologies, through the application of their biomimetic features to the etiology of neurodegenerative diseases. This article reviews the microdevices for nervous system-on-a-chip platform incorporated with neurobiology and microtechnology, including microfluidic designs that are biomimetic to the entire nervous system. The emulation of both neurodegenerative disorders and neural stem cell behavior patterns in micro-platforms is also provided, which can be used as a basis to construct nervous system-on-a-chip.

Quad-phased data mining modeling for dementia diagnosis

BACKGROUND

The number of people with dementia is increasing along with people's ageing trend worldwide. Therefore, there are various researches to improve a dementia diagnosis process in the field of computer-aided diagnosis (CAD) technology. The most significant issue is that the evaluation processes by physician which is based on medical information for patients and questionnaire from their guardians are time consuming, subjective and prone to error. This problem can be solved by an overall data mining modeling, which subsidizes an intuitive decision of clinicians.

METHODS

Therefore, in this paper we propose a quad-phased data mining modeling consisting of 4 modules. In Proposer Module, significant diagnostic criteria are selected that are effective for diagnostics. Then in Predictor Module, a model is constructed to predict and diagnose dementia based on a machine learning algorism. To help clinical physicians understand results of the predictive model better, in Descriptor Module, we interpret causes of diagnostics by profiling patient groups. Lastly, in Visualization Module, we provide visualization to effectively explore characteristics of patient groups.

RESULTS

The proposed model is applied for CREDOS study which contains clinical data collected from 37 university-affiliated hospitals in republic of Korea from year 2005 to 2013.

CONCLUSIONS

This research is an intelligent system enabling intuitive collaboration between CAD system and physicians. And also, improved evaluation process is able to effectively reduce time and cost consuming for clinicians and patients.

Prediction of in-hospital mortality after ruptured abdominal aortic aneurysm repair using an artificial neural network

OBJECTIVE

Ruptured abdominal aortic aneurysm (rAAA) carries a high mortality rate, even with prompt transfer to a medical center. An artificial neural network (ANN) is a computational model that improves predictive ability through pattern recognition while continually adapting to new input data. The goal of this study was to effectively use ANN modeling to provide vascular surgeons a discriminant adjunct to assess the likelihood of in-hospital mortality on a pending rAAA admission using easily obtainable patient information from the field.

METHODS

Of 332 total patients from a single institution from 1998 to 2013 who had attempted rAAA repair, 125 were reviewed for preoperative factors associated with in-hospital mortality; 108 patients received an open operation, and 17 patients received endovascular repair. Five variables were found significant on multivariate analysis (P < .05), and four of these five (preoperative shock, loss of consciousness, cardiac arrest, and age) were modeled by multiple logistic regression and an ANN. These predictive models were compared against the Glasgow Aneurysm Score. All models were assessed by generation of receiver operating characteristic curves and actual vs predicted outcomes plots, with area under the curve and Pearson r(2) value as the primary measures of discriminant ability.

RESULTS

Of the 125 patients, 53 (42%) did not survive to discharge. Five preoperative factors were significant (P < .05) independent predictors of in-hospital mortality in multivariate analysis: advanced age, renal disease, loss of consciousness, cardiac arrest, and shock, although renal disease was excluded from the models. The sequential accumulation of zero to four of these risk factors progressively increased overall mortality rate, from 11% to 16% to 44% to 76% to 89% (age ≥ 70 years considered a risk factor). Algorithms derived from multiple logistic regression, ANN, and Glasgow Aneurysm Score models generated area under the curve values of 0.85 ± 0.04, 0.88 ± 0.04 (training set), and 0.77 ± 0.06 and Pearson r(2) values of .36, .52 and .17, respectively. The ANN model represented the most discriminant of the three.

CONCLUSIONS

An ANN-based predictive model may represent a simple, useful, and highly discriminant adjunct to the vascular surgeon in accurately identifying those patients who may carry a high mortality risk from attempted repair of rAAA, using only easily definable preoperative variables. Although still requiring external validation, our model is available for demonstration at https://redcap.vanderbilt.edu/surveys/?s=NN97NM7DTK.

The use of artificial neural networks to predict delayed discharge and readmission in enhanced recovery following laparoscopic colorectal cancer surgery

BACKGROUND

Artificial neural networks (ANNs) can be used to develop predictive tools to enable the clinical decision-making process. This study aimed to investigate the use of an ANN in predicting the outcomes from enhanced recovery after colorectal cancer surgery.

METHODS

Data were obtained from consecutive colorectal cancer patients undergoing laparoscopic surgery within the enhanced recovery after surgery (ERAS) program between 2002 and 2009 in a single center. The primary outcomes assessed were delayed discharge and readmission within a 30-day period. The data were analyzed using a multilayered perceptron neural network (MLPNN), and a prediction tools were created for each outcome. The results were compared with a conventional statistical method using logistic regression analysis.

RESULTS

A total of 275 cancer patients were included in the study. The median length of stay was 6 days (range 2-49 days) with 67 patients (24.4 %) staying longer than 7 days. Thirty-four patients (12.5 %) were readmitted within 30 days. Important factors predicting delayed discharge were related to failure in compliance with ERAS, particularly with the postoperative elements in the first 48 h. The MLPNN for delayed discharge had an area under a receiver operator characteristic curve (AUROC) of 0.817, compared with an AUROC of 0.807 for the predictive tool developed from logistic regression analysis. Factors predicting 30-day readmission included overall compliance with the ERAS pathway and receiving neoadjuvant treatment for rectal cancer. The MLPNN for readmission had an AUROC of 0.68.

CONCLUSIONS

These results may plausibly suggest that ANN can be used to develop reliable outcome predictive tools in multifactorial intervention such as ERAS. Compliance with ERAS can reliably predict both delayed discharge and 30-day readmission following laparoscopic colorectal cancer surgery.

Model-independent evaluation of tumor markers and a logistic-tree approach to diagnostic decision support

Sensitivity and specificity of using individual tumor markers hardly meet the clinical requirement. This challenge gave rise to many efforts, e.g., combing multiple tumor markers and employing machine learning algorithms. However, results from different studies are often inconsistent, which are partially attributed to the use of different evaluation criteria. Also, the wide use of model-dependent validation leads to high possibility of data overfitting when complex models are used for diagnosis. We propose two model-independent criteria, namely, area under the curve (AUC) and Relief to evaluate the diagnostic values of individual and multiple tumor markers, respectively. For diagnostic decision support, we propose the use of logistic-tree which combines decision tree and logistic regression. Application on a colorectal cancer dataset shows that the proposed evaluation criteria produce results that are consistent with current knowledge. Furthermore, the simple and highly interpretable logistic-tree has diagnostic performance that is competitive with other complex models.

Machine learning applications in cancer prognosis and prediction

Cancer has been characterized as a heterogeneous disease consisting of many different subtypes. The early diagnosis and prognosis of a cancer type have become a necessity in cancer research, as it can facilitate the subsequent clinical management of patients. The importance of classifying cancer patients into high or low risk groups has led many research teams, from the biomedical and the bioinformatics field, to study the application of machine learning (ML) methods. Therefore, these techniques have been utilized as an aim to model the progression and treatment of cancerous conditions. In addition, the ability of ML tools to detect key features from complex datasets reveals their importance. A variety of these techniques, including Artificial Neural Networks (ANNs), Bayesian Networks (BNs), Support Vector Machines (SVMs) and Decision Trees (DTs) have been widely applied in cancer research for the development of predictive models, resulting in effective and accurate decision making. Even though it is evident that the use of ML methods can improve our understanding of cancer progression, an appropriate level of validation is needed in order for these methods to be considered in the everyday clinical practice. In this work, we present a review of recent ML approaches employed in the modeling of cancer progression. The predictive models discussed here are based on various supervised ML techniques as well as on different input features and data samples. Given the growing trend on the application of ML methods in cancer research, we present here the most recent publications that employ these techniques as an aim to model cancer risk or patient outcomes.

Can valid and practical risk-prediction or casemix adjustment models, including adjustment for comorbidity, be generated from English hospital administrative data (Hospital Episode Statistics)? A national observational study

Background

NHS hospitals collect a wealth of administrative data covering accident and emergency (A&E) department attendances, inpatient and day case activity, and outpatient appointments. Such data are increasingly being used to compare units and services, but adjusting for risk is difficult.

Objectives

To derive robust risk-adjustment models for various patient groups, including those admitted for heart failure (HF), acute myocardial infarction, colorectal and orthopaedic surgery, and outcomes adjusting for available patient factors such as comorbidity, using England’s Hospital Episode Statistics (HES) data. To assess if more sophisticated statistical methods based on machine learning such as artificial neural networks (ANNs) outperform traditional logistic regression (LR) for risk prediction. To update and assess for the NHS the Charlson index for comorbidity. To assess the usefulness of outpatient data for these models.

Main outcome measures

Mortality, readmission, return to theatre, outpatient non-attendance. For HF patients we considered various readmission measures such as diagnosis-specific and total within a year.

Methods

We systematically reviewed studies comparing two or more comorbidity indices. Logistic regression, ANNs, support vector machines and random forests were compared for mortality and readmission. Models were assessed using discrimination and calibration statistics. Competing risks proportional hazards regression and various count models were used for future admissions and bed-days.

Results

Our systematic review and empirical analysis suggested that for general purposes comorbidity is currently best described by the set of 30 Elixhauser comorbidities plus dementia. Model discrimination was often high for mortality and poor, or at best moderate, for other outcomes, for examplec = 0.62 for readmission andc = 0.73 for death following stroke. Calibration was often good for procedure groups but poorer for diagnosis groups, with overprediction of low risk a common cause. The machine learning methods we investigated offered little beyond LR for their greater complexity and implementation difficulties. For HF, some patient-level predictors differed by primary diagnosis of readmission but not by length of follow-up. Prior non-attendance at outpatient appointments was a useful, strong predictor of readmission. Hospital-level readmission rates for HF did not correlate with readmission rates for non-HF; hospital performance on national audit process measures largely correlated only with HF readmission rates.

Conclusions

Many practical risk-prediction or casemix adjustment models can be generated from HES data using LR, though an extra step is often required for accurate calibration. Including outpatient data in readmission models is useful. The three machine learning methods we assessed added little with these data. Readmission rates for HF patients should be divided by diagnosis on readmission when used for quality improvement.

Future work

As HES data continue to develop and improve in scope and accuracy, they can be used more, for instance A&E records. The return to theatre metric appears promising and could be extended to other index procedures and specialties. While our data did not warrant the testing of a larger number of machine learning methods, databases augmented with physiological and pathology information, for example, might benefit from methods such as boosted trees. Finally, one could apply the HF readmissions analysis to other chronic conditions.

Funding

The National Institute for Health Research Health Services and Delivery Research programme.

A model to predict limb salvage in severe combat-related open calcaneus fractures

BACKGROUND

Open calcaneus fractures can be limb threatening and almost universally result in some measure of long-term disability. A major goal of initial management in patients with these injuries is setting appropriate expectations and discussing the likelihood of limb salvage, yet there are few tools that assist in predicting the outcome of this difficult fracture pattern.

QUESTIONS/PURPOSES

We developed two decision support tools, an artificial neural network and a logistic regression model, based on presenting data from severe combat-related open calcaneus fractures. We then determined which model more accurately estimated the likelihood of amputation and which was better suited for clinical use.

METHODS

Injury-specific data were collected from wounded active-duty service members who sustained combat-related open calcaneus fractures between 2003 and 2012. One-hundred fifty-five open calcaneus fractures met inclusion criteria. Median followup was 3.5 years (interquartile range: 1.5, 5.1 years), and amputation rate was 44%. We developed an artificial neural network designed to estimate the likelihood of amputation, using information available on presentation. For comparison, a conventional logistic regression model was developed with variables identified on univariate analysis. We determined which model more accurately estimated the likelihood of amputation using receiver operating characteristic analysis. Decision curve analysis was then performed to determine each model's clinical utility.

RESULTS

An artificial neural network that contained eight presenting features resulted in smaller error. The eight features that contributed to the most predictive model were American Society of Anesthesiologist grade, plantar sensation, fracture treatment before arrival, Gustilo-Anderson fracture type, Sanders fracture classification, vascular injury, male sex, and dismounted blast mechanism. The artificial neural network was 30% more accurate, with an area under the curve of 0.8 (compared to 0.65 for logistic regression). Decision curve analysis indicated the artificial neural network resulted in higher benefit across the broadest range of threshold probabilities compared to the logistic regression model and is perhaps better suited for clinical use.

CONCLUSIONS

This report demonstrates an artificial neural network was capable of accurately estimating the likelihood of amputation. Furthermore, decision curve analysis suggested the artificial neural network is better suited for clinical use than logistic regression. Once properly validated, this may provide a tool for surgeons and patients faced with combat-related open calcaneus fractures in which decisions between limb salvage and amputation remain difficult.

Inconsistency in albuminuria predictors in type 2 diabetes: a comparison between neural network and conditional logistic regression

Albuminuria is a sensitive marker to predict future cardiovascular events in patients with type 2 diabetes mellitus. However, current studies only use conventional regression models to discover predictors of albuminuria. We have used 2 different statistical models to predict albuminuria in type 2 diabetes mellitus: a multilayer perception neural network and a conditional logistic regression. Neural network models were used to predict the level of albuminuria in patients with type 2 diabetes mellitus, which include a matched case-control study for the population. For each case, we randomly selected 1 control matched by age and body mass index (BMI). The input variables were sex, duration of diabetes, systolic and diastolic blood pressure, glomerular filtration rate, high-density lipoprotein, low-density lipoprotein, triglyceride, high-density lipoprotein/triglyceride ratio, cholesterol, fasting blood sugar, and glycated hemoglobin. Age and BMI were included only in the neural network model. This model included 4 hidden layers and 1 bias. Relative error of predictions was 0.38% in the training group, 0.52% in the testing group, and 1.20% in the holdout group. The most robust predictors of albuminuria were high-density lipoprotein (21%), cholesterol (14.4%), and systolic blood pressure (9.7%). Using the conditional logistic regression model, glomerular filtration rate, time of onset to diabetes, and sex were significant indicators in the onset of albuminuria. Using a neural network model, we show that high-density lipoprotein is the most important factor in predicting albuminuria in type 2 diabetes mellitus. Our neural network model complements the current risk factor models to improve the care of patients with diabetes.

Artificial neural network for prediction of distant metastasis in colorectal cancer

BACKGROUND AND OBJECTIVES

Artificial neural networks (ANNs) are flexible and nonlinear models which can be used by clinical oncologists in medical research as decision making tools. This study aimed to predict distant metastasis (DM) of colorectal cancer (CRC) patients using an ANN model.

METHODS

The data of this study were gathered from 1219 registered CRC patients at the Research Center for Gastroenterology and Liver Disease of Shahid Beheshti University of Medical Sciences, Tehran, Iran (January 2002 and October 2007). For prediction of DM in CRC patients, neural network (NN) and logistic regression (LR) models were used. Then, the concordance index (C index) and the area under receiver operating characteristic curve (AUROC) were used for comparison of neural network and logistic regression models. Data analysis was performed with R 2.14.1 software.

RESULTS

The C indices of ANN and LR models for colon cancer data were calculated to be 0.812 and 0.779, respectively. Based on testing dataset, the AUROC for ANN and LR models were 0.82 and 0.77, respectively. This means that the accuracy of ANN prediction was better than for LR prediction.

CONCLUSION

The ANN model is a suitable method for predicting DM and in that case is suggested as a good classifier that usefulness to treatment goals.

Prognostic indexes for brain metastases: which is the most powerful?

PURPOSE

The purpose of the present study was to compare the prognostic indexes (PIs) of patients with brain metastases (BMs) treated with whole brain radiotherapy (WBRT) using an artificial neural network. This analysis is important, because it evaluates the prognostic power of each PI to guide clinical decision-making and outcomes research.

METHODS AND MATERIALS

A retrospective prognostic study was conducted of 412 patients with BMs who underwent WBRT between April 1998 and March 2010. The eligibility criteria for patients included having undergone WBRT or WBRT plus neurosurgery. The data were analyzed using the artificial neural network. The input neural data consisted of all prognostic factors included in the 5 PIs (recursive partitioning analysis, graded prognostic assessment [GPA], basic score for BMs, Rotterdam score, and Germany score). The data set was randomly divided into 300 training and 112 testing examples for survival prediction. All 5 PIs were compared using our database of 412 patients with BMs. The sensibility of the 5 indexes to predict survival according to their input variables was determined statistically using receiver operating characteristic curves. The importance of each variable from each PI was subsequently evaluated.

RESULTS

The overall 1-, 2-, and 3-year survival rate was 22%, 10.2%, and 5.1%, respectively. All classes of PIs were significantly associated with survival (recursive partitioning analysis, P < .0001; GPA, P < .0001; basic score for BMs, P = .002; Rotterdam score, P = .001; and Germany score, P < .0001). Comparing the areas under the curves, the GPA was statistically most sensitive in predicting survival (GPA, 86%; recursive partitioning analysis, 81%; basic score for BMs, 79%; Rotterdam, 73%; and Germany score, 77%; P < .001). Among the variables included in each PI, the performance status and presence of extracranial metastases were the most important factors.

CONCLUSION

A variety of prognostic models describe the survival of patients with BMs to a more or less satisfactory degree. Among the 5 PIs evaluated in the present study, GPA was the most powerful in predicting survival. Additional studies should include emerging biologic prognostic factors to improve the sensibility of these PIs.

Which is a more accurate predictor in colorectal survival analysis? Nine data mining algorithms vs. the TNM staging system

OBJECTIVE

Over the past decades, many studies have used data mining technology to predict the 5-year survival rate of colorectal cancer, but there have been few reports that compared multiple data mining algorithms to the TNM classification of malignant tumors (TNM) staging system using a dataset in which the training and testing data were from different sources. Here we compared nine data mining algorithms to the TNM staging system for colorectal survival analysis.

METHODS

Two different datasets were used: 1) the National Cancer Institute's Surveillance, Epidemiology, and End Results dataset; and 2) the dataset from a single Chinese institution. An optimization and prediction system based on nine data mining algorithms as well as two variable selection methods was implemented. The TNM staging system was based on the 7(th) edition of the American Joint Committee on Cancer TNM staging system.

RESULTS

When the training and testing data were from the same sources, all algorithms had slight advantages over the TNM staging system in predictive accuracy. When the data were from different sources, only four algorithms (logistic regression, general regression neural network, bayesian networks, and Naïve Bayes) had slight advantages over the TNM staging system. Also, there was no significant differences among all the algorithms (p>0.05).

CONCLUSIONS

The TNM staging system is simple and practical at present, and data mining methods are not accurate enough to replace the TNM staging system for colorectal cancer survival prediction. Furthermore, there were no significant differences in the predictive accuracy of all the algorithms when the data were from different sources. Building a larger dataset that includes more variables may be important for furthering predictive accuracy.

Using artificial neural networks for open-loop tomography

Modern adaptive optics (AO) systems for large telescopes require tomographic techniques to reconstruct the phase aberrations induced by the turbulent atmosphere along a line of sight to a target which is angularly separated from the guide sources that are used to sample the atmosphere. Multi-object adaptive optics (MOAO) is one such technique. Here, we present a method which uses an artificial neural network (ANN) to reconstruct the target phase given off-axis references sources. We compare our ANN method with a standard least squares type matrix multiplication method and to the learn and apply method developed for the CANARY MOAO instrument. The ANN is trained with a large range of possible turbulent layer positions and therefore does not require any input of the optical turbulence profile. It is therefore less susceptible to changing conditions than some existing methods. We also exploit the non-linear response of the ANN to make it more robust to noisy centroid measurements than other linear techniques.

Computational classifiers for predicting the short-term course of Multiple sclerosis

BACKGROUND

The aim of this study was to assess the diagnostic accuracy (sensitivity and specificity) of clinical, imaging and motor evoked potentials (MEP) for predicting the short-term prognosis of multiple sclerosis (MS).

METHODS

We obtained clinical data, MRI and MEP from a prospective cohort of 51 patients and 20 matched controls followed for two years. Clinical end-points recorded were: 1) expanded disability status scale (EDSS), 2) disability progression, and 3) new relapses. We constructed computational classifiers (Bayesian, random decision-trees, simple logistic-linear regression-and neural networks) and calculated their accuracy by means of a 10-fold cross-validation method. We also validated our findings with a second cohort of 96 MS patients from a second center.

RESULTS

We found that disability at baseline, grey matter volume and MEP were the variables that better correlated with clinical end-points, although their diagnostic accuracy was low. However, classifiers combining the most informative variables, namely baseline disability (EDSS), MRI lesion load and central motor conduction time (CMCT), were much more accurate in predicting future disability. Using the most informative variables (especially EDSS and CMCT) we developed a neural network (NNet) that attained a good performance for predicting the EDSS change. The predictive ability of the neural network was validated in an independent cohort obtaining similar accuracy (80%) for predicting the change in the EDSS two years later.

CONCLUSIONS

The usefulness of clinical variables for predicting the course of MS on an individual basis is limited, despite being associated with the disease course. By training a NNet with the most informative variables we achieved a good accuracy for predicting short-term disability.