Masood A. Khan

A survey on computational intelligence approaches for predictive modeling in prostate cancer

Focus is on computational intelligence methods in prostate cancer predictive modeling.We survey metaheuristic optimisation methods.We review machine learning methods.We consider cancer data of different modalities.We discuss recent advances, challenges and provide future directions. Predictive modeling in medicine involves the development of computational models which are capable of analysing large amounts of data in order to predict healthcare outcomes for individual patients. Computational intelligence approaches are suitable when the data to be modelled are too complex for conventional statistical techniques to process quickly and efficiently. These advanced approaches are based on mathematical models that have been especially developed for dealing with the uncertainty and imprecision which is typically found in clinical and biological datasets. This paper provides a survey of recent work on computational intelligence approaches that have been applied to prostate cancer predictive modeling, and considers the challenges which need to be addressed. In particular, the paper considers a broad definition of computational intelligence which includes metaheuristic optimisation algorithms (also known as nature inspired algorithms), Artificial Neural Networks, Deep Learning, Fuzzy based approaches, and hybrids of these, as well as Bayesian based approaches, and Markov models. Metaheuristic optimisation approaches, such as the Ant Colony Optimisation, Particle Swarm Optimisation, and Artificial Immune Network have been utilised for optimising the performance of prostate cancer predictive models, and the suitability of these approaches are discussed.

When is a bone scan study appropriate in asymptomatic men diagnosed with prostate cancer

AIMSnTo determine when a bone scan investigation is appropriate in asymptomatic men diagnosed with prostate cancer.nnnMETHODSnBetween November 2005 and July 2006, 317 men with prostate cancer underwent a bone scan study; 176 men fulfilled the inclusion criteria. Prostate-specific antigen (PSA) cut-offs as well as univariate and multivariate logistic regression analyses using digital rectal examination finding, biopsy Gleason scores and age were performed to determine when a bone scan study is likely to be of value.nnnRESULTSnOnly 1/61 men (1.6%) with a serum PSA 20 ng/mL had a positive bone scan. However, 2/38 men (4.7%) with a serum PSA 20.1-40.0 ng/mL, 3/20 men (15%) with a serum PSA 40.1-60.0 ng/mL, 7/19 men (36.8%) with a serum PSA 60.1-100.0 ng/mL and 19/38 men (50%) with a serum PSA > 100.0 ng/mL had positive bone scans. Univariate and multivariate logistic regression analyses were uninformative in these groups.nnnCONCLUSIONnBased on our findings, a bone scan is of limited value in asymptomatic prostate cancer patients presenting PSA =or< 20 ng/mL. Therefore, this investigation can be eliminated unless a curative treatment is contemplated. Furthermore, digital rectal examination finding, biopsy Gleason score and age are unhelpful in predicting those who might harbor bone metastasis.

Defining prostate cancer risk before prostate biopsy

Prostate cancer is the most commonly diagnosed cancer in men. At present, patients are selected for prostate biopsy on the basis of age, serum prostate specific antigen (PSA), and prostatic digital rectal examination (DRE) findings. However, due to limitations in the use of PSA and DRE, many patients undergo unnecessary prostate biopsy. A further problem arises as many patients are diagnosed and treated for indolent disease. This review of the literature highlights the strengths and weaknesses of existing methods of prebiopsy risk stratification and evaluates promising serum, urine, and radiologic prostate cancer biomarkers, which may improve risk stratification for prostate biopsy in the future.

Prediction of Pathological Stage in Patients with Prostate Cancer: A Neuro-Fuzzy Model

The prediction of cancer staging in prostate cancer is a process for estimating the likelihood that the cancer has spread before treatment is given to the patient. Although important for determining the most suitable treatment and optimal management strategy for patients, staging continues to present significant challenges to clinicians. Clinical test results such as the pre-treatment Prostate-Specific Antigen (PSA) level, the biopsy most common tumor pattern (Primary Gleason pattern) and the second most common tumor pattern (Secondary Gleason pattern) in tissue biopsies, and the clinical T stage can be used by clinicians to predict the pathological stage of cancer. However, not every patient will return abnormal results in all tests. This significantly influences the capacity to effectively predict the stage of prostate cancer. Herein we have developed a neuro-fuzzy computational intelligence model for classifying and predicting the likelihood of a patient having Organ-Confined Disease (OCD) or Extra-Prostatic Disease (ED) using a prostate cancer patient dataset obtained from The Cancer Genome Atlas (TCGA) Research Network. The system input consisted of the following variables: Primary and Secondary Gleason biopsy patterns, PSA levels, age at diagnosis, and clinical T stage. The performance of the neuro-fuzzy system was compared to other computational intelligence based approaches, namely the Artificial Neural Network, Fuzzy C-Means, Support Vector Machine, the Naive Bayes classifiers, and also the AJCC pTNM Staging Nomogram which is commonly used by clinicians. A comparison of the optimal Receiver Operating Characteristic (ROC) points that were identified using these approaches, revealed that the neuro-fuzzy system, at its optimal point, returns the largest Area Under the ROC Curve (AUC), with a low number of false positives (FPR = 0.274, TPR = 0.789, AUC = 0.812). The proposed approach is also an improvement over the AJCC pTNM Staging Nomogram (FPR = 0.032, TPR = 0.197, AUC = 0.582).

Comet assay measures of DNA damage are predictive of bladder cancer cell treatment sensitivity in vitro and outcome in vivo

Bladder cancer patients suffer significant treatment failure, including high rates of recurrence and poor outcomes for advanced disease. If mechanisms to improve tumour cell treatment sensitivity could be identified and/or if tumour response could be predicted, it should be possible to improve local‐control and survival. Previously, we have shown that radiation‐induced DNA damage, measured by alkaline Comet assay (ACA), correlates bladder cancer cell radiosensitivity in vitro. In this study we first show that modified‐ACA measures of cisplatin and mitomycin‐C‐induced damage also correlate bladder cancer cell chemosensitivity in vitro, with essentially the same rank order for chemosensitivity as for radiosensitivity. Furthermore, ACA studies of radiation‐induced damage in different cell‐DNA substrates (nuclei, nucleoids and intact parent cells) suggest that it is a feature retained in the prepared nucleoids that is responsible for the relative damage sensitivity of bladder cancer cells, suggestive of differences in the organisation of DNA within resistant vs. sensitive cells. Second, we show that ACA analysis of biopsies from bladder tumours reveal that reduced DNA damage sensitivity associates with poorer treatment outcomes, notably that tumours with a reduced damage response show a significant association with local recurrence of non‐invasive disease and that reduced damage response was a better predictor of recurrence than the presence of high‐risk histology in this cohort. In conclusion, this study demonstrates that mechanisms governing treatment‐induced DNA damage are both central to and predictive of bladder cancer cell treatment sensitivity and exemplifies a link between DNA damage resistance and both treatment response and tumour aggression.

Transperineal template prostate biopsies in men with raised PSA despite two previous sets of negative TRUS-guided prostate biopsies

AbstractBackgroundnThe possibility of prostate cancer as a cause for steadily rising PSA despite previously negative transrectal ultrasound (TRUS)-guided prostate biopsies is a major concern. An initial negative TRUS-guided prostate biopsy does not necessarily exclude the presence of clinically significant prostate cancer. We determined the role of transperineal template prostate biopsy (TPTPB) in prostate cancer detection in men with raised PSA despite two previous sets of negative TRUS biopsies.MethodsBetween January 2008 and August 2012, a total of 122 men’s records were reviewed after having 36-core TPTPB following two previous sets of negative TRUS biopsies despite raised PSA. A retrospective record of PSA levels, clinicopathological parameters and histological outcomes was made.ResultsMean age was 63xa0years (range 49–77), and mean PSA was 18.0 (range 2.0–119.0). A total of 71/122 (58xa0%) men were diagnosed with prostate cancer on TPTPB. Of these, 28 (39xa0%), 34 (48xa0%), 5 (7xa0%), and 4 (6xa0%) had Gleason score 6, 7 (3xa0+xa04), 7 (4xa0+xa03), and 9 (4xa0+xa05), respectively. The mean number of positive cores was 7 (range 1–22). Of these, only 15 (21xa0%) had ≤2 cores positive and Gleason score of 6. Of the 51 (42xa0%) men with a negative histology on TPTPB, 11 (22xa0%), 10 (19xa0%), and 30 (59xa0%) had atypical small acinar proliferation, high-grade prostatic intraepithelial neoplasia, or benign pathology.ConclusionTPTPB is associated with a high rate of clinically significant prostate cancer diagnosis (58xa0%) in men with raised PSA despite two previous sets of negative TRUS biopsies.

Incidence and Variables Predicting Gleason Score Up-Grading between Trans-Rectal Ultrasound-Guided Prostate Biopsies and Radical Prostatectomy

Objective: To determine whether Gleason score up-grading is still occurring in men diagnosed with adenocarcinoma of the prostate via extended biopsy regimens, and factors that might predict this. Patients and Methods: Between September 1999 and February 2007, 211 men (age: 42–70 years; mean: 60 years) underwent trans-rectal ultrasound-guided prostate biopsies confirming clinically localized adenocarcinoma followed by radical prostatectomy (RP), within our department. Univariate and multivariate logistic regression (LR) analyses using age, serum PSA, prostate volume, clinical stage and total length of cores taken were performed to determine whether Gleason score up-grading could be predicted. Results: A total of 7/20 (35%), 24/64 (38%) and 36/127 (28%) men with 6, 7–9 and at least 10 core biopsies experienced Gleason score up-grading (p = nonsignificant between the 3 groups). Both univariate and multivariate LR analyses failed to determine any of our variables as a predictor of Gleason score up-grading from biopsy to RP. Conclusion: Despite increasing the number of cores taken at biopsy, in order to improve prostate cancer diagnosis, a substantial percentage of men still experience Gleason score up-grading from biopsy to RP. In addition, we were unable to determine any predicting factors for this up-grading.

Efficacy of a mobile lithotripsy service: A one-year review of 222 patients

Abstract Objective. Extracorporeal shockwave lithotripsy (ESWL) is the management of choice for ureteric and renal stones 20 mm or smaller, with a stone clearance rate of up to 89%. This study determined whether such a high success rate could apply to centres using mobile ESWL, by reviewing the performance at one centre that provides such a service. Material and methods. Between July 2011 and July 2012, 222 patients (median age 51 years, range 18–90 years) underwent one to five sessions of ESWL for ureteric and renal stones (mean size 15 mm, range 4–22 mm). Stone clearance was regarded as residual fragments 2 mm or smaller after completion of sessions. Results. In total, 110 out of 222 patients (49%) were clear of stones. Stones were radiopaque in 198 (89%) and radiolucent in 24 patients (11%), with clearance rates of 48% and 63%, respectively. Regarding size, 36 (16%) were 1–5 mm, 144 (65%) 5–10 mm, 28 (12%) 10–15 mm, eight (4%) 15–20 mm and six (3%) larger than 20 mm, with clearance rates of 61%, 55%, 18%, 13% and 50%, respectively. In total, 173 (78%) were renal stones and 49 (22%) ureteric, with respective clearance rates of 49% and 51%. For kidney stones, 15 (9%) were in the upper, 32 (18%) in the mid, 75 (43%) in the lower pole and 51 (30%) in the pelvis, with clearance rates of 52%, 59%, 49% and 41%; for ureteric stones, 32 (65%) were in the upper, 10 (20%) in the mid and seven (15%) in the lower ureter, with clearance rates of 47%, 70% and 43%, respectively. Conclusion. The performance of mobile ESWL was significantly poorer than expected, and this may be related to a lack of clinical ownership. The authors believe that such a service should be permanently placed on site.

The significance of histological analysis following laser transurethral resection of the prostate

PurposeMonopolar transurethral resection of the prostate is the gold standard for the treatment of benign prostatic hyperplasia. However, due to the associated risks of bleeding and TUR syndrome, laser prostate surgery is gaining popularity. We perform thulium-laser vaporesection of the prostate (TmLRP), where histological samples are generated in every case. We determined postoperative incidence and significance of prostate cancer detection, by retrospective histological examination of our cases.MethodologyBetween October 2006 and August 2012, 223 patients underwent TmLRP by a single surgeon in our institution. With a background of a benign DRE, and no suspicion of CaP, histological results were studied.ResultsMean age was 71xa0years (range 46–91), mean PSA was 4.1xa0ng/mL (range 0.1–20). 4.9% (11/223) had cancer prostate (CaP) diagnosed, with mean PSA of 6.9xa0ng/mL (range 0.7–14). Of these: 91% (10/11) had pT1b disease while 9% (1/11) had pT1a disease. Gleason score was 6 in 28% (3/11); 7 in 36% (4/11); 8 in 18% (2/11); 9 in and 18% (2/11).Conclusion4.9% of patients had unexpected CaP, with significant disease in 4.4% (pT1b) and 3.6% (Gleason score ≥7). Hence, patients should be advised of the small risk of missing significant unsuspected CaP after laser prostatectomy.

PROCEE: a PROstate Cancer Evaluation and Education serious game for African Caribbean men

Purpose n n n n nProstate cancer is the most common cancer diagnosed in men in the UK. Black men are in a higher prostate cancer risk group possibly due to inherent genetic factors. The purpose of this paper is to introduce PROstate Cancer Evaluation and Education (PROCEE), an innovative serious game aimed at providing prostate cancer information and risk evaluation to black African-Caribbean men. n n n n nDesign/methodology/approach n n n n nPROCEE has been carefully co-designed with prostate cancer experts, prostate cancer patients and members of the black African-Caribbean community in order to ensure that it meets the real needs and expectations of the target audience. n n n n nFindings n n n n nDuring the co-design process, the users defined an easy to use and entertaining game which can effectively raise awareness, inform users about prostate cancer and their risk, and encourage symptomatic men to seek medical attention in a timely manner. n n n n nOriginality/value n n n n nDuring focus group evaluations, users embraced the game and emphasised that it can potentially have a positive impact on changing user behaviour among high risk men who are experiencing symptoms and who are reluctant to visit their doctor.