An artificial neural network for prostate cancer staging when serum prostate specific antigen is 10 ng./ml. or less

Comparison of Different Machine Learning Models in Prediction of Postirradiation Recurrence in Prostate Carcinoma Patients

After primary treatment of localized prostate carcinoma (PC), up to a third of patients have disease recurrence. Different predictive models have already been used either for initial stratification of PC patients or to predict disease recurrence. Recently, artificial intelligence has been introduced in the diagnosis and management of PC with a potential to revolutionize this field. The aim of this study was to analyze machine learning (ML) classifiers in order to predict disease progression in the moment of prostate-specific antigen (PSA) elevation during follow-up. The study cohort consisted of 109 PC patients treated with external beam radiotherapy alone or in combination with androgen deprivation therapy. We developed and evaluated the performance of two ML algorithms based on artificial neural networks (ANN) and naïve Bayes (NB). Of all patients, 72.5% was randomly selected for a training set while the remaining patients were used for testing of the models. The presence/absence of disease progression was defined as the output variable. The input variables for models were conducted from the univariate analysis preformed among two groups of patients in the training set. They included two pretreatment variables (UICC stage and Gleason's score risk group) and five posttreatment variables (nadir PSA, time to nadir PSA, PSA doubling time, PSA velocity, and PSA in the moment of disease reevaluation). The area under the receiver operating characteristic curve, sensitivity, specificity, positive predictive value, negative predictive value, and predictive accuracy was calculated to test the models' performance. The results showed that specificity was similar for both models, while NB achieved better sensitivity then ANN (100.0% versus 94.4%). The ANN showed an accuracy of 93.3%, and the matching for NB model was 96.7%. In this study, ML classifiers have shown potential for application in routine clinical practice during follow-up when disease progression was suspected.

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.

Artificial Neural Network: A Method for Prediction of Surgery-Related Pressure Injury in Cardiovascular Surgical Patients

PURPOSE

The aim of this study was to build an artificial neural network (ANN) model for predicting surgery-related pressure injury (SRPI) in cardiovascular surgical patients.

DESIGN

Prospective cohort study.

SUBJECTS AND SETTING

One hundred forty-nine patients who had cardiovascular surgery were included in the study. This study was conducted in a 1000-bed teaching hospital in Eastern China where 250 to 350 cardiac surgeries are performed each year.

METHODS

We performed a prospective cohort study among consecutive patients undergoing cardiovascular surgery between January and December 2015. The ANN model was built based on possible SRPI risk factors. The model performance was tested by a receiver operating characteristic curve and the C-index. A C-index from 0.5 to 0.7 is classified as having low accuracy, 0.7 to 0.9 as having moderate accuracy, and 0.9 to 1.0 as having high accuracy. We also compared the actual SRPI incidences based on the ANN stratification.

RESULTS

Thirty-seven of 147 patients developed SRPIs, yielding an incidence rate of 24.8% (95% CI, 18.1-32.6). The C-index was 0.815, which showed the ANN model had a moderate prediction value for SRPI. According to the ANN model, the SRPI predicting incidence ranged from 6.4% to 67.7%. Surgery-related pressure injury incidences were significantly different among 3 risk groups stratified by the ANN (P < .05).

CONCLUSION

We established an ANN model that provides moderate prediction of SRPI in patients undergoing cardiovascular surgical procedures. Identification and additional associated factors should be incorporated into the ANN model to increase its predictive ability.

Biomarker development, from bench to bedside

This review describes studies performed by our group and other laboratories in the field aimed at development of biomarkers not only for cancer but also for other diseases. The markers covered include tumor-associated trypsin inhibitor (TATI), tumor-associated trypsin (TAT), human chorionic gonadotropin (hCG), prostate-specific antigen (PSA) and their various molecular forms, their biology and diagnostic use. The discovery of TATI was the result of a hypothesis-driven project aimed at finding new biomarkers for ovarian cancer among urinary peptides. TATI has since proved to be a useful prognostic marker for several cancers. Recently, it has been named Serine Peptidase Inhibitor Kazal Type 1 (SPINK1) after being rediscovered by several groups as a tumor-associated peptide by gene expression profiling and proteomic techniques and shown to promote tumor development by stimulating the EGF receptor. To explain why a trypsin inhibitor is strongly expressed in some cancers, research focused on the protease that it inhibited led to the finding of tumor-associated trypsin (TAT). Elevated serum concentrations of TAT-2 were found in some cancer types, but fairly high background levels of pancreatic trypsinogen-2 limited the use of TAT-2 for cancer diagnostics. However, trypsinogen-2 and its complex with α1-protease inhibitor proved to be very sensitive and specific markers for pancreatitis. Studies on hCG were initiated by the need to develop more rapid and sensitive pregnancy tests. These studies showed that serum from men and non-pregnant women contains measurable concentrations of hCG derived from the pituitary. Subsequent development of assays for the subunits of hCG showed that the β subunit of hCG (hCGβ) is expressed at low concentrations by most cancers and that it is a strong prognostic marker. These studies led to the formation of a working group for standardization of hCG determinations and the development of new reference reagents for several molecular forms of hCG. The preparation of intact hCG has been adopted as the fifth international standard by WHO. Availability of several well-defined forms of hCG made it possible to characterize the epitopes of nearly 100 monoclonal antibodies. This will facilitate design of immunoassays with pre-defined specificity. Finally, the discovery of different forms of immunoreactive PSA in serum from a prostate cancer patient led to identification of the complex between PSA and α1-antichymotrypsin, and the use of assays for free and total PSA in serum for improved diagnosis of prostate cancer. Epitope mapping of PSA antibodies and establishment of PSA standards has facilitated establishment well-standardized assays for the various forms of PSA.

Predicting Survival in Patients With Brain Metastases Treated With Radiosurgery Using Artificial Neural Networks

BACKGROUND:

Artificial neural networks (ANNs) excel at analyzing challenging data sets and can be exceptional tools for decision support in clinical environments. The present study pilots the use of ANNs for determining prognosis in neuro-oncology patients.

OBJECTIVE:

To determine whether ANNs perform better at predicting 1-year survival in a group of patients with brain metastasis compared with traditional predictive tools.

METHODS:

ANNs were trained on a multi-institutional data set of radiosurgery patients to predict 1-year survival on the basis of several input factors. A single ANN, an ensemble of 5 ANNs, and logistic regression analyses were compared for efficacy. Sensitivity analysis was used to identify important variables in the ANN model.

RESULTS:

A total of 196 patients were divided up into training, testing, and validation data sets consisting of 98, 49, and 49 patients, respectively. Patients surviving at 1 year tended to be female (P = .001) and of good performance status (P = .01) and to have favorable primary tumor histology (P = .001). The pooled voting of 5 ANNs performed significantly better than the multivariate logistic regression model (P = .02), with areas under the curve of 84% and 75%, respectively. The ensemble also significantly outperformed 2 commonly used prognostic indexes. Primary tumor subtype and performance status were identified on sensitivity analysis to be the most important variables for the ANN.

CONCLUSION:

ANNs outperform traditional statistical tools and scoring indexes for predicting individual patient prognosis. Their facile implementation, robustness in the presence of missing data, and ability to continuously learn make them excellent choices for use in complicated clinical environments.

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.

A comparison of models for predicting sperm retrieval before microdissection testicular sperm extraction in men with nonobstructive azoospermia

PURPOSE

We developed an artificial neural network and nomogram using readily available clinical features to model the chance of identifying sperm with microdissection testicular sperm extraction by readily available preoperative clinical parameters for men with nonobstructive azoospermia.

MATERIALS AND METHODS

We reviewed the records of 1,026 men who underwent microdissection testicular sperm extraction. Patient age, follicle-stimulating hormone level, testicular volume, history of cryptorchidism, Klinefelter syndrome and presence of varicocele were included in the models. For the artificial neural network the data set was divided randomly into a training set (75%) and a test set (25%) with n1/n2 cross validation used to evaluate model accuracy, and then modeled with a neural computational system. In addition, a nomogram with calibration plots was developed to predict sperm retrieval with microdissection testicular sperm extraction. We compared these models to logistic regression.

RESULTS

The ROC area for the neural computational system in the test set was 0.641. The neural network correctly predicted the outcome in 152 of the 256 test set patients (59.4%). The nomogram AUC was 0.59 and adequately calibrated. Multivariable logistic regression demonstrated patient age, history of Klinefelter syndrome and cryptorchidism to be significant predictors of sperm retrieval (p <0.05). However, follicle-stimulating hormone and testicular volume were not significant by internal validation.

CONCLUSIONS

We modeled a combination of well described preoperative clinical parameters to predict sperm retrieval using a neural computational system and nomogram with acceptable predictive values. The generalizability of these findings requires external validation.

Pre-operative prediction of advanced prostatic cancer using clinical decision support systems: accuracy comparison between support vector machine and artificial neural network

Objective

The purpose of the current study was to develop support vector machine (SVM) and artificial neural network (ANN) models for the pre-operative prediction of advanced prostate cancer by using the parameters acquired from transrectal ultrasound (TRUS)-guided prostate biopsies, and to compare the accuracies between the two models.

Materials and Methods

Five hundred thirty-two consecutive patients who underwent prostate biopsies and prostatectomies for prostate cancer were divided into the training and test groups (n = 300 versus n = 232). From the data in the training group, two clinical decision support systems (CDSSs-[SVM and ANN]) were constructed with input (age, prostate specific antigen level, digital rectal examination, and five biopsy parameters) and output data (the probability for advanced prostate cancer [> pT3a]). From the data of the test group, the accuracy of output data was evaluated. The areas under the receiver operating characteristic (ROC) curve (AUC) were calculated to summarize the overall performances, and a comparison of the ROC curves was performed (p < 0.05).

Results

The AUC of SVM and ANN is 0.805 and 0.719, respectively (p = 0.020), in the pre-operative prediction of advanced prostate cancer.

Conclusion

The performance of SVM is superior to ANN in the pre-operative prediction of advanced prostate cancer.

Use of artificial neural networks in the management of antenatally diagnosed ureteropelvic junction obstruction

BACKGROUND

In this study, an artificial neural network (ANN) based system has been developed specifically to help in the management of antenatally diagnosed uretero-pelvic junction (UPJ) obstruction.

METHODS

A total of 53 infants with antenatally detected hydronephrosis caused by UPJ obstruction were included in this study. A neural network was developed with the help of a commercially available software package. The patients' age and sex, renal pelvic diameter, laterality, split renal function and presence of renal scar on radionuclide scan, follow-up times, urine culture results and the presence of symptomatic infections were used as variables. These data were also entered into a statistical software package and linear regression analysis was done.

RESULTS

During the follow-up period, 36 children were observed, and the remaining 17 renal units underwent pyeloplasty. The average sensitivity of the ANN model in predicting the outcome was found to be 92% in the training group and 75% in the validation and test groups. In linear regression, none of the predictors were found to be statistically significant.

INTERPRETATION

In this study, we have demonstrated that the use of ANNs in antenatally diagnosed UPJ obstruction can help the clinician in making treatment decisions, and thus can be useful in daily clinical practice.

Epidemiology and statistical methods in prediction of patient outcome

Substantial gaps exist in the data of the assessment of risk and prognosis that limit our understanding of the complex mechanisms that contribute to the greatest cancer epidemic, prostate cancer, of our time. This report was prepared by an international multidisciplinary committee of the World Health Organization to address contemporary issues of epidemiology and statistical methods in prostate cancer, including a summary of current risk assessment methods and prognostic factors. Emphasis was placed on the relative merits of each of the statistical methods available. We concluded that: 1. An international committee should be created to guide the assessment and validation of molecular biomarkers. The goal is to achieve more precise identification of those who would benefit from treatment. 2. Prostate cancer is a predictable disease despite its biologic heterogeneity. However, the accuracy of predicting it must be improved. We expect that more precise statistical methods will supplant the current staging system. The simplicity and intuitive ease of using the current staging system must be balanced against the serious compromise in accuracy for the individual patient. 3. The most useful new statistical approaches will integrate molecular biomarkers with existing prognostic factors to predict conditional life expectancy (i.e. the expected remaining years of a patient's life) and take into account all-cause mortality.

The value of an artificial neural network in the decision-making for prostate biopsies

PURPOSE

In majority of patients who are subjected to prostate biopsies, no prostate cancer (PCa) is found. It is important to prevent unnecessary biopsies since serious complications may occur. An artificial neural network (ANN) may be able to predict the risk of the presence of PCa.

METHODS

Included were all patients, who underwent transrectal ultrasound-guided prostate biopsies between June 2006 and June 2007 with a total PSA (tPSA) level between 2 and 20 microg/l. The patients were divided into two groups according to their tPSA level (2-10 microg/l and 10-20 microg/l). The ANN Prostataclass of the Universitätsklinikum Charité in Berlin was used. The predictions of the ANN were compared to the pathology results of the biopsies.

RESULTS

Overall 165 patients were included. PCa was diagnosed in 53 patients, whereas the ANN predicted "no risk" in 19 of these patients (36%). The ANN output receiver operator characteristic (ROC) plots for the range of tPSA 2-10 microg/l and tPSA 10-20 microg/l showed an area under the curve (AUC) of 63 and 88% for the initial biopsy group, versus 69 and 57%, respectively, for the repeat biopsy group.

CONCLUSIONS

The ANN resulted in a false negative rate of 36%, missing PCa in 19 patients. For use in an outpatient-clinical setting, this ANN is insufficient to predict the risk of presence of PCa reliably.

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).

Application of artificial neural networks for decision support in medicine

The emergence of drug resistant pathogens can reduce the efficacy of drugs commonly used to treat infectious diseases. Human immunodeficiency virus (HIV) is particularly sensitive to drug selection pressure, rapidly evolving into drug resistant variants on exposure to anti-HIV drugs. Over 200 mutations within the genetic material of HIV have been shown to be associated with drug resistance to date, and complex mutational patterns have been found in HIV isolates from infected patients exposed to multiple antiretroviral drugs. Genotyping is commonly used in clinical practice as a tool to identify drug resistance mutations in HIV from individual patients. This information is then used to help guide the choice of future therapy for patients whose drug regimen is failing because of the development of drug resistant HIV. Many sets of rules and algorithms are available to predict loss of susceptibility to individual antiretroviral drugs from genotypic data. Although this approach has been helpful, the interpretation of genotypic data remains challenging. We describe here the development and application of ANN models as alternative tools for the interpretation of HIV genotypic drug resistance data. A large amount of clinical and virological data, from around 30,000 patients treated with antiretroviral drugs, has been collected by the HIV Resistance Response Database Initiative (RDI, www.hivrdi.org) in a centralized database. Treatment change episodes (TCEs) have been extracted from these data and used along with HIV drug resistance mutations as the basic input variables to train ANN models. We performed a series of analyses that have helped define the following: (1) the reliability of ANN predictions for HIV patients receiving routine clinical care; (2) the utility of ANN models to identify effective treatments for patients failing therapy; (3) strategies to increase the accuracy of ANN predictions; and (4) performance of ANN models in comparison to the rules-based methods currently in use.

Classification models for early detection of prostate cancer

We investigate the performance of different classification models and their ability to recognize prostate cancer in an early stage. We build ensembles of classification models in order to increase the classification performance. We measure the performance of our models in an extensive cross-validation procedure and compare different classification models. The datasets come from clinical examinations and some of the classification models are already in use to support the urologists in their clinical work.

Prognostic accuracy of an artificial neural network in patients undergoing radical cystectomy for bladder cancer: a comparison with logistic regression analysis

OBJECTIVE

To compare the prognostic performance of an artificial neural network (ANN) with that of standard logistic regression (LR), in patients undergoing radical cystectomy for bladder cancer.

PATIENTS AND METHODS

From February 1982 to February 1994, 369 evaluable patients with non-metastatic bladder cancer had pelvic lymph node dissection and radical cystectomy for either stage Ta-T1 (any grade) tumour not responding to intravesical therapy, with or with no carcinoma in situ, or stage T2-T4 tumour. LR analysis based on 12 variables was used to identify predictors of overall 5-year survival, and the ANN model was developed to predict the same outcome. The LR analysis, based on statistically significant predictors, and the ANN model were the compared for their accuracy in predicting survival.

RESULTS

The median age of the patients was 63 years, and overall 201 of them died. The tumour stage and nodal involvement (both P<0.001) were the only statistically independent predictors of overall 5-year survival on LR analysis. Based on these variables, LR had a sensitivity and specificity for predicting survival of 68.4% and 82.8%, respectively; corresponding values for the ANN were 62.7% and 86.1%. For LR and ANN, the positive predictive values were 78.6% and 76.2%, and the negative predictive values were 73.9% and 76.5%, respectively. The index of diagnostic accuracy was 75.9% for LR and 76.4% for ANN.

CONCLUSIONS

The ANN accurately predicted the survival of patients undergoing radical cystectomy for bladder cancer and had a prognostic performance comparable with that of LR. As ANNs are based on easy-to-use software that can identify nonlinear interactions between variables, they might become the preferred tool for predicting outcome.

Potential impact of advanced clinical information technology on cancer care in 2015

New clinical information technologies now sporadically available will soon be in routine clinical use, bringing many changes to all phases of the cancer care continuum. For example, new technologies such as: (1) The next generation Internet; (2) Real-time clinical decision support systems; (3) Off-line, population-based systems; (4) Large, integrated, individual patient-level phenotypic and genotypic databases with intelligent data mining capabilities; (5) Wireless, invasive and non-invasive physiologic monitoring devices; (6) Natural Language Processing (NLP) systems; and (7) Mathematical models of complex biological systems all have the potential to impact significantly the provision of cancer care throughout its continuum. While new information management and communication techniques and technologies will reduce many of the inefficiencies and inaccuracies of our present systems, there will be an equal, and potentially far more dangerous, set of unintended consequences. Informatics investigators, cancer specialists, and health system administrators must focus on the study of what is working and what is not, as well as, on development and testing of the new clinical information management and communication technologies, if we are to be ready for the future.

Computer technology in detection and staging of prostate carcinoma: A review

After two decades of increasing interest and research activity, computer-assisted diagnostic approaches are reaching the stage where more routine deployment in clinical practice is becoming a possibility [Kruppinski, E.A., 2004. Computer-aided detection in clinical environment: Benefits and challenges for radiologists. Radiology 231, 7-9]. This is particularly the case in the analysis of mammographic images [Helvie, M.A., Hadjiiski, L., Makariou, E., Chan, H.P., Petrick, N., Sahiner, B., Lo, S.C., Freedman, M., Adler, D., Bailey, J., Blane, C., Hoff, D., Hunt, K., Joynt, L., Klein, K., Paramagul, C., Patterson, S.K., Roubidoux, M.A., 2004. Sensitivity of noncommercial computer-aided detection system for mammographic breast cancer detection: pilot clinical trial. Radiology 231, 208-214] and in the detection of pulmonary nodules [Reeves, A.P., Kostis, W.J., 2000. Computer-aided diagnosis for lung cancer. Radiol. Clin. North Am. 38, 497-509]. However, similar approaches can be applied more widely with the promise of increasing clinical utility in other areas. We review how computer-aided approaches may be applied in the diagnosis and staging of prostatic cancer. The current status of computer technology is reviewed, covering artificial neural networks for detection and staging, computerised biopsy simulation and computer-assisted analysis of ultrasound and magnetic resonance images.

Little or No Residual Prostate Cancer at Radical Prostatectomy: Vanishing Cancer or Switched Specimen?

With more vigilant screening for prostate cancer, there has been an associated increase in patients with little or no residual cancer at radical prostatectomy after an initial diagnosis of minute cancer on needle biopsy. This raises a critical question as to whether the biopsy and subsequent radical prostatectomy in these patients are from the same patient. We used PCR-based microsatellite marker analysis to perform identity test in 46 men (35 with minute cancer and 11 with no residual cancer). Of them, 41 were interpretable, including 31 with minute cancer and 10 with no residual cancer. All 31 interpretable cases with minute cancer showed match between the initial biopsy and radical prostatectomy specimens. Nine of the 10 interpretable cases with no residual cancer showed match and 1 showed mismatch. The remaining 5 cases (4 with minute cancer and 1 with no residual cancer) were considered uninterpretable due to technical problems. The initial biopsy of the mismatched case had high-grade cancer (Gleason score 4 + 4 = 8) measuring 9.6 mm in length with perineural invasion. Our results confirm that, in most cases of "vanishing cancer" in radical prostatectomy specimens, it reflects a chance sampling of a minute cancer and not a switch in specimens. However, specimen switch can rarely occur, and if there is high grade or a lot of cancer on the biopsy with no or very minimal cancer in the radical prostatectomy specimen, one should evaluate for patient identity.

Relating biopsy and clinical variables to radical prostatectomy findings: Can insignificant and advanced prostate cancer be predicted in a screening population?

OBJECTIVES

To assess the capacity of several clinical and needle biopsy pathologic parameters to predict insignificant and advanced prostate carcinoma (CaP) in radical prostatectomy tissue from men enrolled in a prostate-specific antigen screening program.

METHODS

We captured multiple clinical variables and measures of needle biopsy tumor extent from 152 men with Stage T1c CaP with a mean of six biopsy cores who were treated with radical prostatectomy. Insignificant CaP was defined as a tumor volume of less than 0.5 cm(3) that was organ confined with a Gleason score less than 7. Advanced CaP was defined by a formula that combined the Gleason score, pathologic stage, and margin status. Bivariate and logistic regression analyses were used to identify variables predictive of either insignificant or advanced CaP.

RESULTS

Of the cases of CaP, 25.7% were pathologically insignificant, and 14.5% were pathologically advanced. The best model for predicting insignificant CaP was less than 10% tumor as the greatest percentage of carcinoma in any core and a biopsy Gleason score of less than 7, yielding a sensitivity of 76.9% and specificity of 75.2%. For predicting advanced CaP, the best model was a total biopsy length of CaP greater than 3 mm, Gleason high-grade pattern 4 or 5 disease, perineural invasion in the biopsy, and more than one in six biopsy cores containing CaP, yielding a sensitivity of 13.6% and specificity of 100%.

CONCLUSIONS

The prediction of insignificant and advanced CaP on an individual basis in patients from a prostate-specific antigen screening study is a challenging problem. However, several histopathologic features of CaP in needle biopsy tissue contain useful information about the severity of disease.

Can we improve the prediction of stone-free status after extracorporeal shock wave lithotripsy for ureteral stones? A neural network or a statistical model?

PURPOSE

We evaluated whether an artificial neural network (ANN) can improve the prediction of stone-free status after extracorporeal shock wave lithotripsy (ESWL) (Dornier Medical Systems, Inc., Marietta, Georgia) for ureteral stones compared to a logistic regression (LR) model.

MATERIALS AND METHODS

Between February 1989 and December 1998, 984 patients with ureteral stones, including 780 males and 204 females with a mean age +/- SD of 40.85 +/- 10.33 years, were treated with ESWL. Stone-free status at 3 months was determined by urinary tract plain x-ray and excretory urography. Of all patients 919 (93.3%) were free of stones. The impact of 10 factors on stone-free status was studied using an LR model and ANN. These factors were patient age and sex, renal anatomy, stone location, side, number, length and width, whether stones were de novo or recurrent, and stent use. An LR model was constructed and ANN was trained on 688 randomly selected patients (70%) to predict stone-free status at 3 months. The 10 factors were used as covariates in the LR model and as input parameters to ANN. Performance of the trained net and developed logistic model was evaluated in the remaining 296 patients (30%), who served as the test set. The sensitivity (percent of correctly predicted stone-free cases), specificity (percent of correctly predicted nonstonefree cases), positive predictive value, overall accuracy and average classification rate of the 2 techniques were compared. Relevant variables influencing the construction of the 2 models were compared.

RESULTS

Evaluating the performance of the LR and ANN models on the test set revealed a sensitivity of 100% and 77.9%, a specificity of 0.0% and 75%, a positive predictive value of 93.2% and 97.2%, an overall accuracy of 93.2% and 77.7%, and an average classification rate of 50% and 76.5%, respectively. LR failed to predict any nonstone free cases. LR and ANN identified stone location and stent use as important factors in determining the outcome, while ANN also identified stone length and width as influential factors.

CONCLUSIONS

ANN and LR could predict adequately those who would be stone-free after ESWL for ureteral stones. The neural network has a higher ability to predict those who fail to respond to ESWL.

Artificial neural networks for predictive modeling in prostate cancer

Artificial neural networks (ANNs) represent a relatively new methodology for predictive modeling in medicine. ANNs, a form of artificial intelligence loosely based on the brain, have a demonstrated ability to learn complex and subtle relationships between variables in medical applications. In contrast with traditional statistical techniques, ANNs are capable of automatically resolving these relationships without the need for a priori assumptions about the nature of the interactions between variables. As with any technique, ANNs have limitations and potential drawbacks. This article provides an overview of the theoretical basis of ANNs, how they function, their strengths and limitations, and examples of how ANNs have been used to develop predictive models for the management of prostate cancer.

Comparison of Accuracy Between the Partin Tables Of 1997 and 2001 to Predict Final Pathological Stage in Clinically Localized Prostate Cancer

PURPOSE

We validated externally the predictive accuracy of the 2001 Partin tables and compared the 1997 and 2001 versions.

MATERIALS AND METHODS

We used ROC derived AUC to test the predictive accuracy of organ confinement (OC), extraprostatic extension (ECE), seminal vesicle invasion (SVI) and lymph node involvement (LNI) of 1997 and 2001 Partin tables derived probabilities. These probabilities were defined by the pretreatment clinical stage, serum prostate specific antigen and biopsy Gleason grade of 2,139 patients treated with radical prostatectomy for clinically localized prostate cancer.

RESULTS

OC, ECE, SVI and LNI were noted in 63.5%, 23.1%, 10.5% and 2.9% of cases, respectively. AUC of the 2001 tables was 0.787, 0.766, 0.775 and 0.790, for OC, ECE, SVI and LNI, respectively. These values were virtually the same as the respective 1997 Partin table AUC values, namely 0.784, 0.728, 0.791 and 0.799.

CONCLUSIONS

This external validation of the 2001 Partin tables confirms good predictive accuracy of the updated tables. However, predictive accuracy in this external validation data set of 2,139 European men is virtually the same as that of the original 1997 tables. Therefore, a transition from the 1997 tables to the updated 2001 version does not appear warranted unless superior accuracy is demonstrated in other external cohorts.