A survey on deep learning applied to medical images: from simple artificial neural networks to generative models

Deep learning techniques, in particular generative models, have taken on great importance in medical image analysis. This paper surveys fundamental deep learning concepts related to medical image generation. It provides concise overviews of studies which use some of the latest state-of-the-art models from last years applied to medical images of different injured body areas or organs that have a disease associated with (e.g., brain tumor and COVID-19 lungs pneumonia). The motivation for this study is to offer a comprehensive overview of artificial neural networks (NNs) and deep generative models in medical imaging, so more groups and authors that are not familiar with deep learning take into consideration its use in medicine works. We review the use of generative models, such as generative adversarial networks and variational autoencoders, as techniques to achieve semantic segmentation, data augmentation, and better classification algorithms, among other purposes. In addition, a collection of widely used public medical datasets containing magnetic resonance (MR) images, computed tomography (CT) scans, and common pictures is presented. Finally, we feature a summary of the current state of generative models in medical image including key features, current challenges, and future research paths.

A systematic review of the applications of Expert Systems (ES) and machine learning (ML) in clinical urology

BACKGROUND

Testing a hypothesis for 'factors-outcome effect' is a common quest, but standard statistical regression analysis tools are rendered ineffective by data contaminated with too many noisy variables. Expert Systems (ES) can provide an alternative methodology in analysing data to identify variables with the highest correlation to the outcome. By applying their effective machine learning (ML) abilities, significant research time and costs can be saved. The study aims to systematically review the applications of ES in urological research and their methodological models for effective multi-variate analysis. Their domains, development and validity will be identified.

METHODS

The PRISMA methodology was applied to formulate an effective method for data gathering and analysis. This study search included seven most relevant information sources: WEB OF SCIENCE, EMBASE, BIOSIS CITATION INDEX, SCOPUS, PUBMED, Google Scholar and MEDLINE. Eligible articles were included if they applied one of the known ML models for a clear urological research question involving multivariate analysis. Only articles with pertinent research methods in ES models were included. The analysed data included the system model, applications, input/output variables, target user, validation, and outcomes. Both ML models and the variable analysis were comparatively reported for each system.

RESULTS

The search identified n = 1087 articles from all databases and n = 712 were eligible for examination against inclusion criteria. A total of 168 systems were finally included and systematically analysed demonstrating a recent increase in uptake of ES in academic urology in particular artificial neural networks with 31 systems. Most of the systems were applied in urological oncology (prostate cancer = 15, bladder cancer = 13) where diagnostic, prognostic and survival predictor markers were investigated. Due to the heterogeneity of models and their statistical tests, a meta-analysis was not feasible.

CONCLUSION

ES utility offers an effective ML potential and their applications in research have demonstrated a valid model for multi-variate analysis. The complexity of their development can challenge their uptake in urological clinics whilst the limitation of the statistical tools in this domain has created a gap for further research studies. Integration of computer scientists in academic units has promoted the use of ES in clinical urological research.

Evaluation of prediction models for the staging of prostate cancer

Background

There are dilemmas associated with the diagnosis and prognosis of prostate cancer which has lead to over diagnosis and over treatment. Prediction tools have been developed to assist the treatment of the disease.

Methods

A retrospective review was performed of the Irish Prostate Cancer Research Consortium database and 603 patients were used in the study. Statistical models based on routinely used clinical variables were built using logistic regression, random forests and k nearest neighbours to predict prostate cancer stage. The predictive ability of the models was examined using discrimination metrics, calibration curves and clinical relevance, explored using decision curve analysis. The N = 603 patients were then applied to the 2007 Partin table to compare the predictions from the current gold standard in staging prediction to the models developed in this study.

Results

30% of the study cohort had non organ-confined disease. The model built using logistic regression illustrated the highest discrimination metrics (AUC = 0.622, Sens = 0.647, Spec = 0.601), best calibration and the most clinical relevance based on decision curve analysis. This model also achieved higher discrimination than the 2007 Partin table (ECE AUC = 0.572 & 0.509 for T1c and T2a respectively). However, even the best statistical model does not accurately predict prostate cancer stage.

Conclusions

This study has illustrated the inability of the current clinical variables and the 2007 Partin table to accurately predict prostate cancer stage. New biomarker features are urgently required to address the problem clinician’s face in identifying the most appropriate treatment for their patients. This paper also demonstrated a concise methodological approach to evaluate novel features or prediction models.

Early diagnosis of complex diseases by molecular biomarkers, network biomarkers, and dynamical network biomarkers

Many studies have been carried out for early diagnosis of complex diseases by finding accurate and robust biomarkers specific to respective diseases. In particular, recent rapid advance of high-throughput technologies provides unprecedented rich information to characterize various disease genotypes and phenotypes in a global and also dynamical manner, which significantly accelerates the study of biomarkers from both theoretical and clinical perspectives. Traditionally, molecular biomarkers that distinguish disease samples from normal samples are widely adopted in clinical practices due to their ease of data measurement. However, many of them suffer from low coverage and high false-positive rates or high false-negative rates, which seriously limit their further clinical applications. To overcome those difficulties, network biomarkers (or module biomarkers) attract much attention and also achieve better performance because a network (or subnetwork) is considered to be a more robust form to characterize diseases than individual molecules. But, both molecular biomarkers and network biomarkers mainly distinguish disease samples from normal samples, and they generally cannot ensure to identify predisease samples due to their static nature, thereby lacking ability to early diagnosis. Based on nonlinear dynamical theory and complex network theory, a new concept of dynamical network biomarkers (DNBs, or a dynamical network of biomarkers) has been developed, which is different from traditional static approaches, and the DNB is able to distinguish a predisease state from normal and disease states by even a small number of samples, and therefore has great potential to achieve "real" early diagnosis of complex diseases. In this paper, we comprehensively review the recent advances and developments on molecular biomarkers, network biomarkers, and DNBs in particular, focusing on the biomarkers for early diagnosis of complex diseases considering a small number of samples and high-throughput data (or big data). Detailed comparisons of various types of biomarkers as well as their applications are also discussed.

Use of genetic programming, logistic regression, and artificial neural nets to predict readmission after coronary artery bypass surgery

As many as 14 % of patients undergoing coronary artery bypass surgery are readmitted within 30 days. Readmission is usually the result of morbidity and may lead to death. The purpose of this study is to develop and compare statistical and genetic programming models to predict readmission. Patients were divided into separate Construction and Validation populations. Using 88 variables, logistic regression, genetic programs, and artificial neural nets were used to develop predictive models. Models were first constructed and tested on the Construction populations, then validated on the Validation population. Areas under the receiver operator characteristic curves (AU ROC) were used to compare the models. Two hundred and two patients (7.6 %) in the 2,644 patient Construction group and 216 (8.0 %) of the 2,711 patient Validation group were re-admitted within 30 days of CABG surgery. Logistic regression predicted readmission with AU ROC = .675 ± .021 in the Construction group. Genetic programs significantly improved the accuracy, AU ROC = .767 ± .001, p < .001). Artificial neural nets were less accurate with AU ROC = 0.597 ± .001 in the Construction group. Predictive accuracy of all three techniques fell in the Validation group. However, the accuracy of genetic programming (AU ROC = .654 ± .001) was still trivially but statistically non-significantly better than that of the logistic regression (AU ROC = .644 ± .020, p = .61). Genetic programming and logistic regression provide alternative methods to predict readmission that are similarly accurate.

Artificial neural networks and prostate cancer--tools for diagnosis and management

Artificial neural networks (ANNs) are mathematical models that are based on biological neural networks and are composed of interconnected groups of artificial neurons. ANNs are used to map and predict outcomes in complex relationships between given 'inputs' and sought-after 'outputs' and can also be used find patterns in datasets. In medicine, ANN applications have been used in cancer diagnosis, staging and recurrence prediction since the mid-1990s, when an enormous effort was initiated, especially in prostate cancer detection. Modern ANNs can incorporate new biomarkers and imaging data to improve their predictive power and can offer a number of advantages as clinical decision making tools, such as easy handling of distribution-free input parameters. Most importantly, ANNs consider nonlinear relationships among input data that cannot always be recognized by conventional analyses. In the future, complex medical diagnostic and treatment decisions will be increasingly based on ANNs and other multivariate models.

Machine learning for improved pathological staging of prostate cancer: a performance comparison on a range of classifiers

OBJECTIVES

Prediction of prostate cancer pathological stage is an essential step in a patient's pathway. It determines the treatment that will be applied further. In current practice, urologists use the pathological stage predictions provided in Partin tables to support their decisions. However, Partin tables are based on logistic regression (LR) and built from US data. Our objective is to investigate a range of both predictive methods and of predictive variables for pathological stage prediction and assess them with respect to their predictive quality based on U.K. data.

METHODS AND MATERIAL

The latest version of Partin tables was applied to a large scale British dataset in order to measure their performances by mean of concordance index (c-index). The data was collected by the British Association of Urological Surgeons (BAUS) and gathered records from over 1700 patients treated with prostatectomy in 57 centers across UK. The original methodology was replicated using the BAUS dataset and evaluated using concordance index. In addition, a selection of classifiers, including, among others, LR, artificial neural networks and Bayesian networks (BNs) was applied to the same data and compared with each other using the area under the ROC curve (AUC). Subsets of the data were created in order to observe how classifiers perform with the inclusion of extra variables. Finally a local dataset prepared by the Aberdeen Royal Infirmary was used to study the effect on predictive performance of using different variables.

RESULTS

Partin tables have low predictive quality (c-index=0.602) when applied on UK data for comparison on patients with organ confined and extra prostatic extension conditions, patients at the two most frequently observed pathological stages. The use of replicate lookup tables built from British data shows an improvement in the classification, but the overall predictive quality remains low (c-index=0.610). Comparing a range of classifiers shows that BNs generally outperform other methods. Using the four variables from Partin tables, naive Bayes is the best classifier for the prediction of each class label (AUC=0.662 for OC). When two additional variables are added, the results of LR (0.675), artificial neural networks (0.656) and BN methods (0.679) are overall improved. BNs show higher AUCs than the other methods when the number of variables raises

CONCLUSION

The predictive quality of Partin tables can be described as low to moderate on U.K. data. This means that following the predictions generated by Partin tables, many patients would received an inappropriate treatment, generally associated with a deterioration of their quality of life. In addition to demographic differences between U.K. and the original U.S. population, the methodology and in particular LR present limitations. BN represents a promising alternative to LR from which prostate cancer staging can benefit. Heuristic search for structure learning and the inclusion of more variables are elements that further improve BN models quality.

Computer-aided interpretation approach for optical tomographic images

A computer-aided interpretation approach is proposed to detect rheumatic arthritis (RA) in human finger joints using optical tomographic images. The image interpretation method employs a classification algorithm that makes use of a so-called self-organizing mapping scheme to classify fingers as either affected or unaffected by RA. Unlike in previous studies, this allows for combining multiple image features, such as minimum and maximum values of the absorption coefficient for identifying affected and not affected joints. Classification performances obtained by the proposed method were evaluated in terms of sensitivity, specificity, Youden index, and mutual information. Different methods (i.e., clinical diagnostics, ultrasound imaging, magnet resonance imaging, and inspection of optical tomographic images), were used to produce ground truth benchmarks to determine the performance of image interpretations. Using data from 100 finger joints, findings suggest that some parameter combinations lead to higher sensitivities, while others to higher specificities when compared to single parameter classifications employed in previous studies. Maximum performances are reached when combining the minimum/maximum ratio of the absorption coefficient and image variance. In this case, sensitivities and specificities over 0.9 can be achieved. These values are much higher than values obtained when only single parameter classifications were used, where sensitivities and specificities remained well below 0.8.

Image-based clinical decision support for transrectal ultrasound in the diagnosis of prostate cancer: comparison of multiple logistic regression, artificial neural network, and support vector machine

PURPOSE

We developed a multiple logistic regression model, an artificial neural network (ANN), and a support vector machine (SVM) model to predict the outcome of a prostate biopsy, and compared the accuracies of each model.

METHOD

One thousand and seventy-seven consecutive patients who had undergone transrectal ultrasound (TRUS)-guided prostate biopsy were enrolled in the study. Clinical decision models were constructed from the input data of age, digital rectal examination findings, prostate-specific antigen (PSA), PSA density (PSAD), PSAD in transitional zone, and TRUS findings. The patients were divided into the training and test groups in a randomized fashion. Areas under the receiver operating characteristic (ROC) curve (AUC, Az) were calculated to summarize the overall performance of each decision model for the task of prostate cancer prediction.

RESULTS

The Az values of the ROC curves for the use of multiple logistic regression analysis, ANN, and the SVM were 0.768, 0.778, and 0.847, respectively. Pairwise comparison of the ROC curves determined that the performance of the SVM was superior to that of the ANN or the multiple logistic regression model.

CONCLUSION

Image-based clinical decision support models allow patients to be informed of the actual probability of having a prostate cancer.

Staging of prostate cancer using automatic feature selection, sampling and Dempster-Shafer fusion

A novel technique of automatically selecting the best pairs of features and sampling techniques to predict the stage of prostate cancer is proposed in this study. The problem of class imbalance, which is prominent in most medical data sets is also addressed here. Three feature subsets obtained by the use of principal components analysis (PCA), genetic algorithm (GA) and rough sets (RS) based approaches were also used in the study. The performance of under-sampling, synthetic minority over-sampling technique (SMOTE) and a combination of the two were also investigated and the performance of the obtained models was compared. To combine the classifier outputs, we used the Dempster-Shafer (DS) theory, whereas the actual choice of combined models was made using a GA. We found that the best performance for the overall system resulted from the use of under sampled data combined with rough sets based features modeled as a support vector machine (SVM).

A [-2]proPSA-based artificial neural network significantly improves differentiation between prostate cancer and benign prostatic diseases

BACKGROUND

The aim of this study was to combine the new automated Access [-2]proPSA (p2PSA) assay with a percent free PSA (%fPSA) based artificial neural network (ANN) or logistic regression (LR) model to enhance discrimination between patients with prostate cancer (PCa) and with no evidence of malignancy (NEM) and to detect aggressive PCa.

METHODS

Sera from 311 PCa patients and 275 NEM patients were measured with the p2PSA, total PSA (tPSA) and free PSA (fPSA) assays on Access immunoassay technology (Beckman Coulter, Fullerton, CA) within the 0-30 ng/ml tPSA range. Four hundred seventy-five patients (264 PCa, 211 NEM) had a tPSA of 2-10 ng/ml. LR models and leave-one-out (LOO) ANN models with Bayesian regularization by using tPSA, %fPSA, p2PSA/fPSA (%p2PSA), age and prostate volume were constructed and compared by receiver-operating characteristic (ROC) curve analysis.

RESULTS

The ANN and LR model each utilizing %p2PSA, %fPSA, tPSA and age, but without prostate volume, reached the highest AUCs (0.85 and 0.84) and best specificities (ANN: 62.1% and 45.5%; LR: 53.1% and 41.2%) compared with tPSA (22.7% and 11.4%) and %fPSA (45.5% and 26.1%) at 90% and 95% sensitivity. The %p2PSA furthermore distinguished better than tPSA and %fPSA between pT2 and pT3, and Gleason sum <7 and >or=7 PCa.

CONCLUSIONS

The automated p2PSA assay offers a new tool to improve PCa detection, and especially aggressive PCa detection. Incorporation of %p2PSA into an ANN and LR model further enhances the diagnostic accuracy to differentiate between malignant and non-malignant prostate diseases.

Multiparameter classifications of optical tomographic images

This research study explores the combined use of more than one parameter derived from optical tomographic images to increase diagnostic accuracy which is measured in terms of sensitivity and specificity. Parameters considered include, for example, smallest or largest absorption or scattering coefficients or the ratios thereof in an image region of interest. These parameters have been used individually in a previous study to determine if a finger joint is affected or not affected by rheumatoid arthritis. To combine these parameters in the analysis we employ here a vector quantization based classification method called Self-Organizing Mapping (SOM). This method allows producing multivariate ROC-curves from which sensitivity and specificities can be determined. We found that some parameter combinations can lead to higher sensitivities whereas others to higher specificities when compared to singleparameter classifications employed in previous studies. The best diagnostic accuracy, in terms of highest Youden index, was achieved by combining three absorption parameters [maximum(micro a), minimum(micro a), and the ratio of minimum(micro a) and maximum(micro a)], which result in a sensitivity of 0.78, a specificity of 0.76, a Youden index of 0.54, and an area under the curve (AUC) of 0.72. These values are higher than for previously reported single parameter classifications with a best sensitivity and specificity of 0.71, a Youden index of 0.41, and an AUC of 0.66.

Genetic classification of severe early childhood caries by use of subtracted DNA fragments from Streptococcus mutans

Streptococcus mutans is one of several members of the oral indigenous biota linked with severe early childhood caries (S-ECC). Because most humans harbor S. mutans, but not all manifest disease, it has been proposed that the strains of S. mutans associated with S-ECC are genetically distinct from those found in caries-free (CF) children. The objective of this study was to identify common DNA fragments from S. mutans present in S-ECC but not in CF children. Using suppressive subtractive hybridization, we found a number of DNA fragments (biomarkers) present in 88 to 95% of the S-ECC S. mutans strains but not in CF S. mutans strains. We then applied machine learning techniques including support vector machines and neural networks to identify the biomarkers with the most predictive power for disease status, achieving a 92% accurate classification of the strains as either S-ECC or CF associated. The presence of these gene fragments in 90 to 100% of the 26 S-ECC isolates tested suggested their possible functional role in the pathogenesis of S. mutans associated with dental caries.

Comparing the predictive value of neural network models to logistic regression models on the risk of death for small-cell lung cancer patients

Cancer is one of the leading causes of mortality in the developed world, and prognostic assessment of cancer patients is indispensable in medical care. Medical researchers are accustomed to using regression models to predict patient outcomes. Neural networks have been proposed as an alternative with great potential. Nonetheless, empirical evidence remains lacking to support the application of this technique as the appropriate method to investigate cancer prognosis. Utilizing data on patients from two National Cancer Institute of Canada clinical trials, we compared predictive accuracy of neural network models and logistic regression models on risk of death of limited-stage small-cell lung cancer patients. Our results suggest that neural network and logistic regression models have similar predictive accuracy. The distributions of individual predicted probabilities are very similar. On occasion, however, the prediction pairs are quite different, suggesting that they do not always give the same interpretations of the same variables.

Identification of patients with low-risk for aneuploidy: comparative discriminatory models using linear and machine-learning classifiers in prostate cancer

BACKGROUND

Prostate needle biopsy (PNB) ploidy status has proven utility to predict adverse outcomes after prostatectomy. We sought to develop models to predict ploidy status using clinicopathologic variables.

METHODS

We identified a cohort of 169 patients with a diagnosis of prostatic adenocarcinoma on PNB, and estimated ploidy status (determined using Feulgen stained biopsy tissue) using four predictors, including age, prebiopsy PSA, highest Gleason score (GS), and the percentage of involvement by carcinoma at the biopsy site with the highest GS (PCARBX). Logistic regression (LR), Neural Network (NN), and CART classifiers were constructed.

RESULTS

Univariate analyses revealed all four predictors to be significantly associated with ploidy status. On multivariable analyses, LR identified a 2-parameter model, including GS and PCARBX that had a significant ability to predict ploidy status with a 74% and 75% correct classification rate (CCR), respectively. Using the same variables, CART and NN yielded similar CCRs of 70.4%. Within GS = 6 cohort, the CART model classified over 90% of biopsies as diploid when patients had a PCARBX < 55% and a log(PSA) < 1.7.

CONCLUSIONS

Our study demonstrates that models using GS and PCARBX are able to predict PNB ploidy status with acceptable accuracy. While machine learning classifier-derived models yield similar accuracy as LR-derived models, the latter methodology has the distinct advantage of being applicable in future datasets to estimate case-specific predictions. This information may be useful in identifying potentially aneuploid patients, who can then be targeted for more aggressive therapy.

Improved prostate cancer detection with a human kallikrein 11 and percentage free PSA-based artificial neural network

Human kallikrein 11 (hK11) was evaluated in a percentage free PSA-based artificial neural network (ANN) to reduce unnecessary prostate biopsies. Serum samples from 357 patients with (n=132) and without (n=225) prostate cancer (PCa) were analyzed and ANN models were constructed and compared to all parameters. The discriminatory power of hK11 was lower than that of PSA, but receiver operator characteristic (ROC) analyses demonstrated significantly larger areas under the curves for the ANN compared to all other parameters. ANNs with hK11 may lead to a further reduction in unnecessary prostate biopsies, especially when analyzing patients with less than 15% free PSA.

Role of transrectal ultrasonography in the prediction of prostate cancer: artificial neural network analysis

OBJECTIVE

The purpose of this study was to evaluate the diagnostic performance of an artificial neural network (ANN) model with and without transrectal ultrasonographic (TRUS) data.

METHODS

Six hundred eighty-four consecutive patients who had undergone TRUS-guided prostate biopsy from May 2003 to January 2005 were enrolled. We constructed 2 ANN models. One (ANN_1) incorporated patient age, digital rectal examination findings, prostate-specific antigen (PSA) level, PSA density, transitional zone volume, and PSA density in the transitional zone as input data, whereas the other (ANN_2) was constructed with the above and TRUS findings as input data. The performances of these 2 ANN models according to PSA levels (group A, 0-4 ng/mL; group B, 4-10 ng/mL; and group C, >10 ng/mL) were evaluated using receiver operating characteristic analysis.

RESULTS

Of the 684 patients who underwent prostate biopsy, 214 (31.3%) were confirmed to have prostate cancer; of 137 patients with positive digital rectal examination results, 60 (43.8%) were confirmed to have prostate cancer; and of 131 patients with positive TRUS findings, 93 (71%) were confirmed to have prostate cancer. In groups A, B, and C, the AUCs for ANN_1 were 0.738, 0.753, and 0.774, respectively; the AUCs for ANN_2 were 0.859, 0.797, and 0.894. In all groups, ANN_2 showed better accuracy than ANN_1 (P < .05).

CONCLUSIONS

According to receiver operating characteristic analysis, ANN with TRUS findings was found to be more accurate than ANN without. We conclude that TRUS findings should be included as an input data component in ANN models used to diagnose prostate cancer.

Three new serum markers for prostate cancer detection within a percent free PSA-based artificial neural network

BACKGROUND

We aimed to evaluate the value of macrophage inhibitory cytokine 1 (MIC-1), human kallikrein 11 (hK11) migration inhibitor factor (MIF) in comparison to prostate-specific antigen (PSA) and % fPSA and also to develop a % fPSA-based ANN with the new input factors to determine whether these additional markers can further eliminate unnecessary prostate biopsies.

METHODS

Serum samples from 371 patients with prostate cancer (PCa, n=135) or benign prostate hyperplasia (BPH, n=236) within the PSA range 0.5-20 microg/L were analyzed for total PSA, free PSA, MIC-1, hK11, and MIF. 'Leave one out' ANN models with these variables and prostate volume were constructed and compared to logistic regression (LR) and all single parameters.

RESULTS

The discriminatory power of MIC-1, hK11, and MIF was less than that for PSA despite significant differences in BPH compared to PCa patients. At 90% and 95% sensitivity, the artificial neural networks (ANNs) were only significantly better than % fPSA if prostate volume was included.

CONCLUSIONS

ANNs with the novel input factors of MIC-1, MIF, and/or hK11 and additional use of prostate volume demonstrated significant advantage compared with % fPSA and tPSA and may lead to a reduction in unnecessary prostate biopsies.

THE PERCENT OF BIOPSY CORES POSITIVE FOR CANCER IS A PREDICTOR OF ADVANCED PATHOLOGICAL STAGE AND POOR CLINICAL OUTCOMES IN PATIENTS TREATED WITH RADICAL PROSTATECTOMY

PURPOSE

We examined if the percent of positive biopsies is associated with features of biologically aggressive prostate cancer, biochemical progression and development of distant metastases in patients undergoing radical prostatectomy (RP).

MATERIALS AND METHODS

Multivariate analyses of preoperative features in 605 consecutive patients who underwent RP for clinically localized disease were evaluated to determine the association between the percent positive biopsy cores (PosBx), pathological stage and grade, and biochemical progression following RP. The percent of PosBx cores was defined using the formula, (number of positive biopsy cores/total number of biopsy cores) x 100.

RESULTS

The mean number of biopsy cores and percent PosBx cores +/- SE was 8.8 +/- 6.0 and 31.4 +/-21.1, respectively. Higher percent PosBx was significantly associated with higher preoperative prostate specific antigen (PSA), extracapsular extension, seminal vesicle invasion, positive surgical margins, higher final Gleason sum, lymphovascular invasion, perineural invasion and metastases to regional lymph nodes. On multivariate analyses adjusted for the effects of standard preoperative features percent PosBx was associated with nonorgan confined disease, seminal vesicle invasion and biochemical progression after surgery (p = 0.049, 0.050 and 0.006, respectively). Percent PosBx retained its independent association with PSA progression after adjustment for the effects of postoperative pathological features (p = 0.015). Higher percent PosBx was associated with shorter PSA doubling time after PSA progression, and an increased risk of distant metastases and overall mortality (p = 0.039, 0.001 and 0.018, respectively).

CONCLUSIONS

Percent PosBx is associated with established pathological features, biochemical progression, distant metastases and overall death in patients who undergo RP for clinically localized disease. Percent PosBx should be included in preoperative predictive models for prognosticating outcomes after primary treatment and it may assist in selecting patients for inclusion in neoadjuvant and/or adjuvant therapy trials.

Clinical application of artificial neural network (ANN) modeling to predict pharmacokinetic parameters of severely ill patients

Artificial neural network (ANN) modeling was used to evaluate the pharmacokinetics of aminoglycosides (arbekacin sulfate and amikacin sulfate) in severely ill patients. The plasma level was predicted by ANN modeling using parameters related to the severity of the patient's condition and the predictive performance was shown to be better than could be achieved using multiple regression analysis. These results indicate that there is a non-linear relationship between the pharmacokinetics of aminoglycosides and the severity of the patient's condition, and this should be taken into account when determining the dose for severely ill patients. Patients whose plasma levels are likely to fall below the effective level can be identified by ANN modeling with a predictive sensitivity and specificity superior to multivariate logistic regression analysis. The predictable range should be inferred from the data structure before the modeling in order to improve the predictive performance. The volume of distribution (Vd) in the normal range was weakly predicted by ANN modeling from the patients' data. Prediction of clearance by ANN modeling was poorer than that obtained from serum creatinine concentration by linear regression analysis. These results suggest that the input-output relationship (linear or non-linear) should be taken into account in selecting the modeling method. Linear modeling can give better predictive performance for linear systems and non-linear modeling can give better predictive performance for non-linear systems. In general, the performance of ANN modeling was superior to linear modeling for PK/PD prediction. For accurate modeling, a predictable range should be inferred from the data structure before the analysis. Restriction of the predictable region, as determined from the data structure, produced an increase in prediction performance. When applying ANN modeling in clinical settings, the predictive performance and predictable region should be investigated in detail to avoid the risk of harm to severely ill patients.