Improving Prostate Cancer (PCa) Classification Performance by Using Three-Player Minimax Game to Reduce Data Source Heterogeneity

2020 ◽  
Vol 39 (10) ◽  
pp. 3148-3158
Author(s):  
Yanan Shao ◽  
Jane Wang ◽  
Brian Wodlinger ◽  
Septimiu E. Salcudean
Keyword(s):  
Prostate Cancer ◽  
Classification Performance ◽  
Data Source ◽  
Minimax Game ◽  
Source Heterogeneity

scholarly journals Textured-Based Deep Learning in Prostate Cancer Classification with 3T Multiparametric MRI: Comparison with PI-RADS-Based Classification

Diagnostics ◽  
2021 ◽  
Vol 11 (10) ◽  
pp. 1785
Author(s):  
Yongkai Liu ◽  
Haoxin Zheng ◽  
Zhengrong Liang ◽  
Qi Miao ◽  
Wayne G. Brisbane ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Deep Learning ◽  
Sensitivity And Specificity ◽  
Predictive Value ◽  
Peripheral Zone ◽  
Significant Degree ◽  
Classification Performance ◽  
Multiparametric Mri ◽  
Superior Performance ◽  
Study Cohort

The current standardized scheme for interpreting MRI requires a high level of expertise and exhibits a significant degree of inter-reader and intra-reader variability. An automated prostate cancer (PCa) classification can improve the ability of MRI to assess the spectrum of PCa. The purpose of the study was to evaluate the performance of a texture-based deep learning model (Textured-DL) for differentiating between clinically significant PCa (csPCa) and non-csPCa and to compare the Textured-DL with Prostate Imaging Reporting and Data System (PI-RADS)-based classification (PI-RADS-CLA), where a threshold of PI-RADS ≥ 4, representing highly suspicious lesions for csPCa, was applied. The study cohort included 402 patients (60% (n = 239) of patients for training, 10% (n = 42) for validation, and 30% (n = 121) for testing) with 3T multiparametric MRI matched with whole-mount histopathology after radical prostatectomy. For a given suspicious prostate lesion, the volumetric patches of T2-Weighted MRI and apparent diffusion coefficient images were cropped and used as the input to Textured-DL, consisting of a 3D gray-level co-occurrence matrix extractor and a CNN. PI-RADS-CLA by an expert reader served as a baseline to compare classification performance with Textured-DL in differentiating csPCa from non-csPCa. Sensitivity and specificity comparisons were performed using Mcnemar’s test. Bootstrapping with 1000 samples was performed to estimate the 95% confidence interval (CI) for AUC. CIs of sensitivity and specificity were calculated by the Wald method. The Textured-DL model achieved an AUC of 0.85 (CI [0.79, 0.91]), which was significantly higher than the PI-RADS-CLA (AUC of 0.73 (CI [0.65, 0.80]); p < 0.05) for PCa classification, and the specificity was significantly different between Textured-DL and PI-RADS-CLA (0.70 (CI [0.59, 0.82]) vs. 0.47 (CI [0.35, 0.59]); p < 0.05). In sub-analyses, Textured-DL demonstrated significantly higher specificities in the peripheral zone (PZ) and solitary tumor lesions compared to the PI-RADS-CLA (0.78 (CI [0.66, 0.90]) vs. 0.42 (CI [0.28, 0.57]); 0.75 (CI [0.54, 0.96]) vs. 0.38 [0.14, 0.61]; all p values < 0.05). Moreover, Textured-DL demonstrated a high negative predictive value of 92% while maintaining a high positive predictive value of 58% among the lesions with a PI-RADS score of 3. In conclusion, the Textured-DL model was superior to the PI-RADS-CLA in the classification of PCa. In addition, Textured-DL demonstrated superior performance in the specificities for the peripheral zone and solitary tumors compared with PI-RADS-based risk assessment.

scholarly journals (Dis)concordance of comorbidity data across ICD codes, medical charts, and self-reports

2019 ◽  
Author(s):  
Abdullah Sheriffdeen ◽  
Jeremy Millar ◽  
Catherine Martin ◽  
Melanie Evans ◽  
Gabriella Tikellis ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Administrative Data ◽  
International Classification Of Diseases ◽  
Medical Chart ◽  
Chart Audit ◽  
Standard Data ◽  
Self Reports ◽  
Classification Of Diseases ◽  
Data Source ◽  
Level Of Agreement

Abstract Background Benchmarking outcomes across settings commonly requires risk-adjustment for co-morbidities that must be derived from extant sources that were designed for other purposes. A question arises as to the extent to which differing available sources for health data will be concordant when inferring the type and severity of co-morbidities, how close these are to the “truth”. We studied the level of concordance for same-patient comorbidity data extracted from administrative data (coded from International Classification of Diseases, Australian modification,10th edition [ICD-10AM]), from the medical chart audit, and data self-reported by men with prostate cancer who had undergone a radical prostatectomy.Methods We included six hospitals (5 public and 1 private) contributing to the Prostate Cancer Outcomes Registry-Victoria (PCOR-Vic) in the study. We listed eligible patients from the PCOR-Vic who underwent a radical prostatectomy between January 2017 and April 2018 for the Health Information Manager in each hospital, who provided each patient’s associated ICD-10AM comorbidity codes. Medical charts were reviewed to extract comorbidities used to generate the Charlson Comorbidity Index. The self-reported comorbidity questionnaire (SCQ) was distributed through PCOR-Vic to eligible men.Results The percentage agreement between the administrative data, medical charts and self-reports ranged from 92% to 99% in the 122 patients (from 217 eligible participants, 56%), who responded to the questionnaire. The prevalence-adjusted bias-adjusted kappa (PABAK) coefficient was from 0.83 to 0.98 for all conditions aside from cancer, reflecting a strong level of agreement for the absence of comorbidities. Conversely, the presence of comorbidities showed a poor level of agreement between data sources. There was concordance on 213/277 (77%) comorbidities when comparing medical charts and administrative data; 102/238 (30%) comorbidities when comparing medical chart and self-reports; and 34/150 (23%) comorbidities when comparing administrative data and self-reports.Conclusion Relying on a single data source to generate comorbidity indices for risk-modelling purposes may fail to capture the reality of a patient’s disease profile. There does not appear to be a ‘gold-standard’ data source for the collection of data on comorbidities.

scholarly journals (Dis)concordance of comorbidity data and cancer status across administrative datasets, medical charts, and self-reports

2020 ◽  
Vol 20 (1) ◽  
Author(s):  
A. Sheriffdeen ◽  
J. L. Millar ◽  
C. Martin ◽  
M. Evans ◽  
G. Tikellis ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Radical Prostatectomy ◽  
Administrative Data ◽  
International Classification Of Diseases ◽  
Chart Audit ◽  
Standard Data ◽  
Self Reports ◽  
Classification Of Diseases ◽  
Icd 10 ◽  
Data Source

Abstract Background Benchmarking outcomes across settings commonly requires risk-adjustment for co-morbidities that must be derived from extant sources that were designed for other purposes. A question arises as to the extent to which differing available sources for health data will be concordant when inferring the type and severity of co-morbidities, how close are these to the “truth”. We studied the level of concordance for same-patient comorbidity data extracted from administrative data (coded from International Classification of Diseases, Australian modification,10th edition [ICD-10 AM]), from the medical chart audit, and data self-reported by men with prostate cancer who had undergone a radical prostatectomy. Methods We included six hospitals (5 public and 1 private) contributing to the Prostate Cancer Outcomes Registry-Victoria (PCOR-Vic) in the study. Eligible patients from the PCOR-Vic underwent a radical prostatectomy between January 2017 and April 2018.Health Information Manager’s in each hospital, provided each patient’s associated administrative ICD-10 AM comorbidity codes. Medical charts were reviewed to extract comorbidity data. The self-reported comorbidity questionnaire (SCQ) was distributed through PCOR-Vic to eligible men. Results The percentage agreement between the administrative data, medical charts and self-reports ranged from 92 to 99% in the 122 patients from the 217 eligible participants who responded to the questionnaire. The presence of comorbidities showed a poor level of agreement between data sources. Conclusion Relying on a single data source to generate comorbidity indices for risk-modelling purposes may fail to capture the reality of a patient’s disease profile. There does not appear to be a ‘gold-standard’ data source for the collection of data on comorbidities.

scholarly journals (Dis)concordance of comorbidity data and cancer status across administrative datasets, medical charts, and self-reports

2020 ◽  
Author(s):  
Abdullah Sheriffdeen ◽  
Jeremy Millar ◽  
Catherine Martin ◽  
Melanie Evans ◽  
Gabriella Tikellis ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Radical Prostatectomy ◽  
Administrative Data ◽  
International Classification Of Diseases ◽  
Chart Audit ◽  
Standard Data ◽  
Self Reports ◽  
Classification Of Diseases ◽  
Data Source ◽  
Single Data

Abstract Background Benchmarking outcomes across settings commonly requires risk-adjustment for co-morbidities that must be derived from extant sources that were designed for other purposes. A question arises as to the extent to which differing available sources for health data will be concordant when inferring the type and severity of co-morbidities, how close these are to the “truth”. We studied the level of concordance for same-patient comorbidity data extracted from administrative data (coded from International Classification of Diseases, Australian modification,10 th edition [ICD-10AM]), from the medical chart audit, and data self-reported by men with prostate cancer who had undergone a radical prostatectomy. Methods We included six hospitals (5 public and 1 private) contributing to the Prostate Cancer Outcomes Registry-Victoria (PCOR-Vic) in the study. We listed eligible patients from the PCOR-Vic who underwent a radical prostatectomy between January 2017 and April 2018 for the Health Information Manager in each hospital, who provided each patient’s associated ICD-10AM comorbidity codes. Medical charts were reviewed to extract comorbidities used to generate the Charlson Comorbidity Index. The self-reported comorbidity questionnaire (SCQ) was distributed through PCOR-Vic to eligible men. Results The percentage agreement between the administrative data, medical charts and self-reports ranged from 92% to 99% in the 122 patients (from 217 eligible participants, 56%), who responded to the questionnaire. The presence of comorbidities showed a poor level of agreement between data sources. Due to a variety of factors, certain conditions were recorded more than others. Conclusion Relying on a single data source to generate comorbidity indices for risk-modelling purposes may fail to capture the reality of a patient’s disease profile. There does not appear to be a ‘gold-standard’ data source for the collection of data on comorbidities.

scholarly journals Combination possibility and deep learning model as clinical decision-aided approach for prostate cancer

2019 ◽  
Vol 26 (2) ◽  
pp. 945-962 ◽  
Author(s):  
Okyaz Eminaga ◽  
Omran Al-Hamad ◽  
Martin Boegemann ◽  
Bernhard Breil ◽  
Axel Semjonow
Keyword(s):  
Prostate Cancer ◽  
Deep Learning ◽  
Gleason Score ◽  
Clinical Decision ◽  
Learning Model ◽  
Classification Performance ◽  
Study Data ◽  
Learning Approaches ◽  
Tumor Stages ◽  
Deep Learning Model

This study aims to introduce as proof of concept a combination model for classification of prostate cancer using deep learning approaches. We utilized patients with prostate cancer who underwent surgical treatment representing the various conditions of disease progression. All possible combinations of significant variables from logistic regression and correlation analyses were determined from study data sets. The combination possibility and deep learning model was developed to predict these combinations that represented clinically meaningful patient’s subgroups. The observed relative frequencies of different tumor stages and Gleason score Gls changes from biopsy to prostatectomy were available for each group. Deep learning models and seven machine learning approaches were compared for the classification performance of Gleason score changes and pT2 stage. Deep models achieved the highest F1 scores by pT2 tumors (0.849) and Gls change (0.574). Combination possibility and deep learning model is a useful decision-aided tool for prostate cancer and to group patients with prostate cancer into clinically meaningful groups.

scholarly journals Deep learning facilitates rapid classification of human and veterinary clinical narratives

2018 ◽  
Author(s):  
Arturo Lopez Pineda ◽  
Oliver J. Bear Don’t Walk ◽  
Guhan R. Venkataraman ◽  
Ashley M. Zehnder ◽  
Sandeep Ayyar ◽  
...  
Keyword(s):  
Public Health ◽  
Medical Records ◽  
Short Term Memory ◽  
Classification Performance ◽  
Clinical Settings ◽  
Strong Positive Correlation ◽  
Training Samples ◽  
Cohort Selection ◽  
Data Source ◽  
Evaluation Metric

ABSTRACTObjectiveCurrently, dedicated tagging staff spend considerable effort assigning clinical codes to patient summaries for public health purposes, and machine-learning automated tagging is bottlenecked by availability of electronic medical records. Veterinary medical records, a largely untapped data source that could benefit both human and non-human patients, could fill the gap.Materials and MethodsIn this retrospective study, we trained long short-term memory (LSTM) recurrent neural networks (RNNs) on 52,722 human and 89,591 veterinary records. We established relevant baselines by training Decision Trees (DT) and Random Forests (RF) on the same data. We finally investigated the effect of merging data across clinical settings and probed model portability.ResultsWe show that the LSTM-RNNs accurately classify veterinary/human text narratives into top-level categories with an average weighted macro F1, score of 0.735/0.675 respectively. The evaluation metric for the LSTM was 7 and 8% higher than that of the DT and RF models respectively. We generally did not find evidence of model portability albeit moderate performance increases in select categories.DiscussionWe see a strong positive correlation between number of training samples and classification performance, which is promising for future efforts. The use of LSTM-RNN models represents a scalable structure that could prove useful in cohort selection, which could in turn better address emerging public health concerns.ConclusionDigitization of human and veterinary health information will continue to be a reality. Our approach is a step forward for these two domains to learn from, and inform, one another.

scholarly journals Classification Models for Early Detection of Prostate Cancer

2008 ◽  
Vol 2008 ◽  
pp. 1-7 ◽  
Author(s):  
Joerg D. Wichard ◽  
Henning Cammann ◽  
Carsten Stephan ◽  
Thomas Tolxdorff
Keyword(s):  
Prostate Cancer ◽  
Early Detection ◽  
Cross Validation ◽  
Early Stage ◽  
Clinical Work ◽  
Classification Performance ◽  
Classification Models ◽  
Validation Procedure

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.

An Ontology Approach for Manufacturing Enterprise Data Warehouses Development

2011 ◽  
Vol 215 ◽  
pp. 77-82 ◽  
Author(s):  
B.Y. Xu ◽  
H.M. Cai ◽  
C. Xie
Keyword(s):  
Data Warehouse ◽  
Data Model ◽  
Development Project ◽  
Meta Data ◽  
Manufacturing Enterprises ◽  
Data Heterogeneity ◽  
Data Source ◽  
Enterprise Data Warehouse ◽  
Source Heterogeneity ◽  
Operational Data

Data warehouse (DW) is a powerful and useful technology for decision making in manufacturing enterprises. Because that the operational data often comes from distributed units for manufacturing enterprises, there exits an urgent need to study on the methods of integrating heterogonous data in data warehouse. In This paper, an ontology approach is proposed to eliminate data source heterogeneity. The approach is based on the exploitation of the application of domain ontology methods in data warehouse design, representing the semantic meanings of the data by ontology at database level and pushing the data as data resources to manufacturing units at data warehouse access level. The foundation of our approach is a meta-data model which consists of data, concept, ontology and resource repositories. The model is used in a shipbuilding enterprise data warehouse development project. The result shows that with the guide of the meta-data model, our ontology approach could eliminate the data heterogeneity.

Sigmoidoscopic findings following radiation for prostate cancer

2001 ◽  
Vol 120 (5) ◽  
pp. A284-A284
Author(s):  
T BOLIN ◽  
A KNEEBONE ◽  
T LARSSON
Keyword(s):  
Prostate Cancer
Sign in / Sign up

Export Citation Format

Share Document