scholarly journals Performance Evaluation of Deep Learning-Based Prostate Cancer Screening Methods in Histopathological Images: Measuring the Impact of the Model’s Complexity on Its Processing Speed

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1122 ◽  
Author(s):  
Lourdes Duran-Lopez ◽  
Juan P. Dominguez-Morales ◽  
Antonio Rios-Navarro ◽  
Daniel Gutierrez-Galan ◽  
Angel Jimenez-Fernandez ◽  
...  

Prostate cancer (PCa) is the second most frequently diagnosed cancer among men worldwide, with almost 1.3 million new cases and 360,000 deaths in 2018. As it has been estimated, its mortality will double by 2040, mostly in countries with limited resources. These numbers suggest that recent trends in deep learning-based computer-aided diagnosis could play an important role, serving as screening methods for PCa detection. These algorithms have already been used with histopathological images in many works, in which authors tend to focus on achieving high accuracy results for classifying between malignant and normal cases. These results are commonly obtained by training very deep and complex convolutional neural networks, which require high computing power and resources not only in this process, but also in the inference step. As the number of cases rises in regions with limited resources, reducing prediction time becomes more important. In this work, we measured the performance of current state-of-the-art models for PCa detection with a novel benchmark and compared the results with PROMETEO, a custom architecture that we proposed. The results of the comprehensive comparison show that using dedicated models for specific applications could be of great importance in the future.

Diagnostics ◽  
2020 ◽  
Vol 10 (11) ◽  
pp. 951
Author(s):  
David J. Winkel ◽  
Christian Wetterauer ◽  
Marc Oliver Matthias ◽  
Bin Lou ◽  
Bibo Shi ◽  
...  

Background: Opportunistic prostate cancer (PCa) screening is a controversial topic. Magnetic resonance imaging (MRI) has proven to detect prostate cancer with a high sensitivity and specificity, leading to the idea to perform an image-guided prostate cancer (PCa) screening; Methods: We evaluated a prospectively enrolled cohort of 49 healthy men participating in a dedicated image-guided PCa screening trial employing a biparametric MRI (bpMRI) protocol consisting of T2-weighted (T2w) and diffusion weighted imaging (DWI) sequences. Datasets were analyzed both by human readers and by a fully automated artificial intelligence (AI) software using deep learning (DL). Agreement between the algorithm and the reports—serving as the ground truth—was compared on a per-case and per-lesion level using metrics of diagnostic accuracy and k statistics; Results: The DL method yielded an 87% sensitivity (33/38) and 50% specificity (5/10) with a k of 0.42. 12/28 (43%) Prostate Imaging Reporting and Data System (PI-RADS) 3, 16/22 (73%) PI-RADS 4, and 5/5 (100%) PI-RADS 5 lesions were detected compared to the ground truth. Targeted biopsy revealed PCa in six participants, all correctly diagnosed by both the human readers and AI. Conclusions: The results of our study show that in our AI-assisted, image-guided prostate cancer screening the software solution was able to identify highly suspicious lesions and has the potential to effectively guide the targeted-biopsy workflow.


2016 ◽  
Vol 3 (3) ◽  
pp. 176 ◽  
Author(s):  
Mara M Epstein ◽  
Daniel M Frendl ◽  
Hassan Fouayzi ◽  
Richard Krajenta ◽  
Benjamin A Rybicki ◽  
...  

2017 ◽  
Vol 35 (6_suppl) ◽  
pp. 211-211
Author(s):  
Marc Dall'Era ◽  
Ralph deVere White ◽  
Danielle Rodgriguez ◽  
Rosemary Donaldson Cress

211 Background: The United States Preventive Services Task Force (USPSTF) recommended against routine PSA based prostate cancer screening in all men in 2012. This led to dramatic reductions in screening and rates of localized disease across all clinical risk groups. We sought to study the impact of this on rates of metastatic disease, specifically by patient race and age. Methods: We analyzed new prostate cancer incidence by stage at diagnosis between 1988-2013 within the Cancer Registry of Greater California. We further stratified cases by four major race/ethnicity groups (non-Hispanic white (NHW), non-Hispanic black (NHB), Hispanic and non-Hispanic Asian/PI (API)) and age. Incidence rates were calculated and compared per 100,000 and age-adjusted to the 2000 US Standard Population. Joinpoint regression was used to detect changes in incidence and to calculate the average percent change (APC). Results: Adjusted rates of remote prostate cancer incidence for NHW men increased slightly in the most recent decade (+0.28%) after steady declines in previous years with the inflection point occurring in 2002, however this was not statistically significant. In contrast, incidence of remote prostate cancer continued to decline for NHB (-2.73%), Hispanic (-2.04%), and API (-1.45%) men. The greatest increase of +1.1% a year since 2002 was observed for NHW men under age 65. The incidence of localized prostate cancer declined for all race/ethnicity groups over the most recent time period and also declined in all age groups. After remaining relatively flat since 1992, incidence of localized prostate cancer among NHW men declined by over 8% per year starting in 2007 compared with a more gradual decline of -3.52% a year since 2000 for NHB, and more recent declines of -14.41% and -16.64% for Hispanic and API men, respectively. Incidence of regional stage cancer also declined in all groups, but less dramatically. Conclusions: Incidence rates of newly metastatic prostate cancer have not significantly changed since PSA screening declined in the US although we noted a slight upward trend primarily for younger, white men since 2002.


2020 ◽  
Vol 8 (4) ◽  
pp. 243-252
Author(s):  
S.O. Olarewaju ◽  
Amudat D. Akinola ◽  
Emmanuel O. Oyekunle ◽  
Sunday C. Adeyemo

Background: Prostate cancer (PCa) is the second leading cause of cancer-related death in the world. This study assessed factors influencing uptake of PCa screening at a popular town in Iseyin, Oyo State.Methods: Relevant information was obtained from 376 participants using pre-tested semi-structured questionnaires. Respondents were selected through multistage sampling technique Statistical tests such as Chisquare, Fisher's exact test and Student's T-test were performed to duly analyse data obtained.Results: Participants' mean age was 50.2 + 8.0 years with a larger (58. 5 %) proportion of men aged 50 and below. Overall summarized scores on poor knowledge and negative attitude were 42.8 % and 44.7 % respectively. Specific knowledge regarding location of prostate, risk factors and symptoms was as low as 21.0%, 26.3% and 37.2%, respectively. Barely 27.9 %, 19.9 % and 16.2 % knew prostate specific antigen test, digital rectal examination and ultrasound as PC screening techniques respectively. Previous uptake of screening was very low being 16.0 %. Individuals with poor knowledge and those who are non-professionals are 3 times (Odds Ratio – 0.295) and 2 times (Odds Ratio – 0.524) respectively less likely to do PCa screening compared with their other counterpartsConclusion: The study showed considerable awareness of prostate cancer. However, comprehensive knowledge of its symptoms and the screening methods was greatly lacking. The need is therefore indicated for an aggressive health promotion intervention designed to increase awareness on PCa screening at the community level. Keywords: prostate cancer, uptake, screening methods


Author(s):  
Ausvydas Patasius ◽  
Giedre Smailyte

Background: The aim of this study was to examine the impact of screening introduction on prostate cancer incidence changes, and changes in stage distribution in Lithuania between 1998–2016. Methods: Age-standardized incidence as well as stage-specific incidence rates were calculated. Joinpoint regression was used to estimate the annual percentage change in the incidence changes by determined stage: Localized, advanced, distant and unknown. Results: Over the study period, a total number of 48,815 new prostate cancer cases was identified. Age-standardized incidence rose from 51.9 per 100,000 in 1998 to 279.3 per 100,000 in 2007 (by 20.3% per year) and then decreased thereafter by 3.8% annually. Highest incidence rates after introduction of prostate specific antigene (PSA)-based screening was found for localized disease, followed by advanced. Incidence of localized disease rose by 38.2% per year until 2007 reaching the highest rate of 284.6 per 100,000, with a subsequent decrease of 5.5% every year thereafter. Advanced stage of disease experienced rise till 2007, and continuous decrease by 11.1% every year thereafter. Incidence of disease with distant metastasis was lowest, and rose till 2003, thereafter incidence significantly decreased by 8.1% every year. Conclusions: To our knowledge, this is the first report of stage migration effect in Lithuania, following the introduction of nationwide PSA-based screening. Prostate cancer screening substantially increased the overall incidence and incidence of localized cancer.


2014 ◽  
Vol 191 (6) ◽  
pp. 1737-1742 ◽  
Author(s):  
Afshin Aslani ◽  
Brian J. Minnillo ◽  
Ben Johnson ◽  
Edward E. Cherullo ◽  
Lee E. Ponsky ◽  
...  

2015 ◽  
Vol 49 (5) ◽  
pp. 762-768 ◽  
Author(s):  
Amy J. Starosta ◽  
George Luta ◽  
Catherine A. Tomko ◽  
Marc D. Schwartz ◽  
Kathryn L. Taylor

2019 ◽  
Vol 25 (1) ◽  
Author(s):  
Rogers A. Enemugwem ◽  
Beatrice A. Eze ◽  
Ukamaka Ejike ◽  
Eme O. Asuquo ◽  
Alafaka Tobin

Abstract Background Worldwide prostate cancer is the second most frequent cause of cancer deaths and is the commonest cancer diagnosed among Nigerian men. Screening techniques can be employed to detect the disease earlier in apparently healthy individuals, and increasing evidence shows that this can decrease morbidity and mortality of the disease. The objective of this study was to assess the knowledge and intention to screen for prostate cancer among men in Obio Akpor Local Government Area, Rivers State, Nigeria. Results Respondents were within the ages of 40 and 75 years. The most frequently reported source of information about prostate cancer screening was the news media 72 (35.0%) and healthcare workers 62 (30.1%). Thirty (14.9%) of the respondents had good knowledge of prostate cancer, while 80 (39.6%) had good knowledge of prostate cancer screening methods. Concerning prostate cancer screening methods, only 47 (23.3%) were able to correctly identify screening methods for prostate cancer. With regards to intention to screen, 104 (51.5%) were willing to be screened for prostate cancer. Conclusion This study showed that the knowledge of prostate cancer was poor with only approximately half of the participants expressing intentions to screen for the disease. This emphasizes the need for the Rivers State Ministry of Health to carry out awareness campaigns on the importance of prostate screening.


Sign in / Sign up

Export Citation Format

Share Document