A Cascaded Deep Learning–Based Artificial Intelligence Algorithm for Automated Lesion Detection and Classification on Biparametric Prostate Magnetic Resonance Imaging

Mehralivand, Sherif; Yang, Dong; Harmon, Stephanie; Xu, Daguang; Xu, Ziyue; Roth, Holger R.; Masoudi, Samira; Sanford, Thomas; Kesani, Deepak; Lay, Nathan; Merino, María J.; Wood, Bradford J.; Pinto, Peter A.; Choyke, Peter L.; Türkbey, Barış

doi:10.1016/j.acra.2021.08.019

Cited by 32 publications

(13 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The advent of deep learning for medical imaging allows realizing stand-alone AI that achieves good to expert level performance in the prediction of segmentation volume and csPCa detection [ 14 , 15 , 45 ]. Deep learning AI models are being incorporated in products that provide human interface software that aims to help improve workflow and reduce diagnostic performance variability [ 45 , 47 , 60 , 71 ]. Moreover, these AI diagnostic models can be used before, during and after radiation therapy.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Artificial intelligence for prostate MRI: open datasets, available applications, and grand challenges

Sunoqrot

Saha²,

Hosseinzadeh³

et al. 2022

Eur Radiol Exp

View full text Add to dashboard Cite

Artificial intelligence (AI) for prostate magnetic resonance imaging (MRI) is starting to play a clinical role for prostate cancer (PCa) patients. AI-assisted reading is feasible, allowing workflow reduction. A total of 3,369 multi-vendor prostate MRI cases are available in open datasets, acquired from 2003 to 2021 in Europe or USA at 3 T (n = 3,018; 89.6%) or 1.5 T (n = 296; 8.8%), 346 cases scanned with endorectal coil (10.3%), 3,023 (89.7%) with phased-array surface coils; 412 collected for anatomical segmentation tasks, 3,096 for PCa detection/classification; for 2,240 cases lesions delineation is available and 56 cases have matching histopathologic images; for 2,620 cases the PSA level is provided; the total size of all open datasets amounts to approximately 253 GB. Of note, quality of annotations provided per dataset highly differ and attention must be paid when using these datasets (e.g., data overlap). Seven grand challenges and commercial applications from eleven vendors are here considered. Few small studies provided prospective validation. More work is needed, in particular validation on large-scale multi-institutional, well-curated public datasets to test general applicability. Moreover, AI needs to be explored for clinical stages other than detection/characterization (e.g., follow-up, prognosis, interventions, and focal treatment).

show abstract

Section: Discussionmentioning

confidence: 99%

“…PROSTATEx [22] is currently the dataset that is most commonly used for development of AI for detection of csPCa (e.g. [45][46][47]).…”

Section: Table 1 Summary Of Prostate Mri Public Datasetsmentioning

confidence: 99%

Artificial intelligence for prostate MRI: open datasets, available applications, and grand challenges

Sunoqrot

Saha²,

Hosseinzadeh³

et al. 2022

Eur Radiol Exp

View full text Add to dashboard Cite

show abstract

“…With an employment of 3D spatial information from MR series, their accuracy for cancer detection was 0.894. Recently, Mehralivand et al [ 100 ] proposed a cascaded DL model for lesion detection and scoring on bpMRI. The model contained a 3-D UNet-based network that automatically detected and segmented prostate MRI lesions, and two 3D residual networks that made 4-class classification to predict PI-RADS categories.…”

Section: Future Opportunitiesmentioning

confidence: 99%

Machine Learning in Prostate MRI for Prostate Cancer: Current Status and Future Opportunities

Lee

Chia

et al. 2022

Diagnostics

View full text Add to dashboard Cite

Advances in our understanding of the role of magnetic resonance imaging (MRI) for the detection of prostate cancer have enabled its integration into clinical routines in the past two decades. The Prostate Imaging Reporting and Data System (PI-RADS) is an established imaging-based scoring system that scores the probability of clinically significant prostate cancer on MRI to guide management. Image fusion technology allows one to combine the superior soft tissue contrast resolution of MRI, with real-time anatomical depiction using ultrasound or computed tomography. This allows the accurate mapping of prostate cancer for targeted biopsy and treatment. Machine learning provides vast opportunities for automated organ and lesion depiction that could increase the reproducibility of PI-RADS categorisation, and improve co-registration across imaging modalities to enhance diagnostic and treatment methods that can then be individualised based on clinical risk of malignancy. In this article, we provide a comprehensive and contemporary review of advancements, and share insights into new opportunities in this field.

show abstract

“…18,19 Another limitation is the lack of application of deep convolutional neural networks (DCCN) for the diagnosis and classification of BPH, though, studies have been conducted on DCCN to differentiate between PCa and BPH. [20][21][22] Wong et al 20 used deep-learning-based networks to detect PCa in transitional zone (TZ) and differentiate it from BPH using T2-weighted (T2W) and apparent diffusion coefficient (ADC) map MR images. Using dataset of 196 patients, the model using ADC alone demonstrated higher sensitivity (0.829) and precision (0.534) as compared to T2W and T2W + ADC models.…”

Section: Strengths Limitations and Areas Of Future Researchmentioning

confidence: 99%

“…Using dataset of 196 patients, the model using ADC alone demonstrated higher sensitivity (0.829) and precision (0.534) as compared to T2W and T2W + ADC models. 20 A similar study was conducted by Hu et al 21 using deep transfer learning (DTL) methods to overcome the disadvantage of smaller 22 Studies can be conducted in the future on similar lines to solely diagnose, classify, as well determine the severity of BPH using radiomics and DCCN. AI studies conducted in the future should also focus more on quality of life, suitability of the various treatment options, as well as the cost of treatment, and come up with common algorithms that can be used universally.…”

Section: Strengths Limitations and Areas Of Future Researchmentioning

confidence: 99%

Current Applications of Artificial Intelligence in Benign Prostatic Hyperplasia

Shah¹,

Naik²,

Hameed³

et al. 2022

Turkish Journal of Urology

View full text Add to dashboard Cite

Artificial intelligence is used in predicting the clinical outcomes before minimally invasive treatments for benign prostatic hyperplasia, to address the insufficient reliability despite multiple assessment parameters, such as flow rates and symptom scores. Various models of artificial intelligence and its contemporary applications in benign prostatic hyperplasia are reviewed and discussed. A search strategy adapted to identify and review the literature on the application of artificial intelligence with a dedicated search string with the following keywords: "Machine Learning," "Artificial Intelligence," AND "Benign Prostate Enlargement" OR "BPH" OR "Benign Prostatic Hyperplasia" was included and categorized. Review articles, editorial comments, and non-urologic studies were excluded. In the present review, 1600 patients were included from 4 studies that used different classifiers such as fuzzy systems, computer-based vision systems, and clinical data mining to study the applications of artificial intelligence in diagnoses and severity prediction and determine clinical factors responsible for treatment response in benign prostatic hyperplasia. The accuracy to correctly diagnose benign prostatic hyperplasia by Fuzzy systems was 90%, while that of computer-based vision system was 96.3%. Data mining achieved sensitivity and specificity of 70% and 50%, respectively, in correctly predicting the clinical response to medical treatment in benign prostatic hyperplasia. Artificial intelligence is gaining attraction in urology, with the potential to improve diagnostics and patient care. The results of artificial intelligence-based applications in benign prostatic hyperplasia are promising but lack generalizability of results. However, in the future, we will see a shift in the clinical paradigm as artificial intelligence applications will find their place in the guidelines and revolutionize the decision-making process.

show abstract

A Cascaded Deep Learning–Based Artificial Intelligence Algorithm for Automated Lesion Detection and Classification on Biparametric Prostate Magnetic Resonance Imaging

Cited by 32 publications

References 18 publications

Artificial intelligence for prostate MRI: open datasets, available applications, and grand challenges

Artificial intelligence for prostate MRI: open datasets, available applications, and grand challenges

Machine Learning in Prostate MRI for Prostate Cancer: Current Status and Future Opportunities

Current Applications of Artificial Intelligence in Benign Prostatic Hyperplasia

Contact Info

Product

Resources

About