The practical implementation of artificial intelligence technologies in medicine

He, Jianxing; Baxter, Sally L.; Xu, Jie; Xu, Jiming; Zhou, Xingtao; Zhang, Kang

doi:10.1038/s41591-018-0307-0

Cited by 1,365 publications

(1,080 citation statements)

References 22 publications

Supporting

Mentioning

1,047

Contrasting

Unclassified

Order By: Relevance

“…However, radiomics and ML methods have not been evaluated in terms of their repeatability, which contributes to reservations of making imaging applications that would use prostate MRI more extensively coupled with machine learning in routine clinical practice. High short‐term repeatability is a prerequisite toward quantitative‐tailored treatment planning and therapy monitoring, a major determining factor of the potential role of ML for prostate MRI in routine clinical practice in the future …”

Section: Introductionmentioning

confidence: 99%

Repeatability of radiomics and machine learning for DWI: Short‐term repeatability study of 112 patients with prostate cancer

Merisaari

Taimen

Shiradkar

et al. 2019

Magnetic Resonance in Med

View full text Add to dashboard Cite

Purpose To evaluate repeatability of prostate DWI‐derived radiomics and machine learning methods for prostate cancer (PCa) characterization. Methods A total of 112 patients with diagnosed PCa underwent 2 prostate MRI examinations (Scan1 and Scan2) performed on the same day. DWI was performed using 12 b‐values (0–2000 s/mm2), post‐processed using kurtosis function, and PCa areas were annotated using whole mount prostatectomy sections. A total of 1694 radiomic features including Sobel, Kirch, Gradient, Zernike Moments, Gabor, Haralick, CoLIAGe, Haar wavelet coefficients, 3D analogue to Laws features, 2D contours, and corner detectors were calculated. Radiomics and 4 feature pruning methods (area under the receiver operator characteristic curve, maximum relevance minimum redundancy, Spearman’s ρ, Wilcoxon rank‐sum) were evaluated in terms of Scan1‐Scan2 repeatability using intraclass correlation coefficient (ICC)(3,1). Classification performance for clinically significant and insignificant PCa with Gleason grade groups 1 versus >1 was evaluated by area under the receiver operator characteristic curve in unseen random 30% data split. Results The ICC(3,1) values for conventional radiomics and feature pruning methods were in the range of 0.28–0.90. The machine learning classifications varied between Scan1 and Scan2 with % of same class labels between Scan1 and Scan2 in the range of 61–81%. Surface‐to‐volume ratio and corner detector‐based features were among the most represented features with high repeatability, ICC(3,1) >0.75, consistently high ranking using all 4 feature pruning methods, and classification performance with area under the receiver operator characteristic curve >0.70. Conclusion Surface‐to‐volume ratio and corner detectors for prostate DWI led to good classification of unseen data and performed similarly in Scan1 and Scan2 in contrast to multiple conventional radiomic features.

show abstract

Section: Introductionmentioning

confidence: 99%

Repeatability of radiomics and machine learning for DWI: Short‐term repeatability study of 112 patients with prostate cancer

Merisaari

Taimen

Shiradkar

et al. 2019

Magnetic Resonance in Med

View full text Add to dashboard Cite

show abstract

“…Situated within the broader context of the implementation of artificial intelligence in the health care system, generative methods can address key concerns regarding data sharing and quality assurance. 39 The current era of "big data" in medicine has the potential to significantly improve patient safety, but concurrently raises major privacy concerns, particularly in the context of data sharing between institutions and the possibility that combining multiple anonymized datasets can allow for re-identification. 22 As the generator network in a GAN has no direct access to the patient images, these models, in combination with differential privacy strategies 40 , can capture the features of a dataset yet be transferred with minimal risk to patient privacy.…”

Section: Discussionmentioning

confidence: 99%

Synthesis of diagnostic quality cancer pathology images

Peng

Farnell

et al. 2020

Preprint

View full text Add to dashboard Cite

Keyboards: cancer, pathology, deep learning, artificial intelligence, quality assurance, education Link to the demo website: http://gan.aimlab.ca/ 1 ABSTRACT Deep learning-based computer vision methods have recently made remarkable breakthroughs in the analysis and classification of cancer pathology images. However, there has been relatively little investigation of the utility of deep neural networks to synthesize medical images. In this study, we evaluated the efficacy of generative adversarial networks (GANs) to synthesize high resolution pathology images of ten histological types of cancer, including five cancer types from The Cancer Genome Atlas (TCGA) and the five major histological subtypes of ovarian carcinoma. The quality of these images was assessed using a comprehensive survey of board-certified pathologists (n = 9) and pathology trainees (n = 6). Our results show that the real and synthetic images are classified by histotype with comparable accuracies, and the synthetic images are visually indistinguishable from real images. Furthermore, we trained deep convolutional neural networks (CNNs) to diagnose the different cancer types and determined that the synthetic images perform as well as additional real images when used to supplement a small training set. These findings have important applications in proficiency testing of medical practitioners and quality assurance in clinical laboratories. Furthermore, training of computer-aided diagnostic systems can benefit from synthetic images where labeled datasets are limited (e.g., rare cancers). We have created a publicly available website where clinicians and researchers can attempt questions from the image survey at http://gan.aimlab.ca/ .

show abstract

“…Therefore, development of software‐based support for PoC decision‐making is crucial for the success of precision medicine. AI might provide such possibilities in the future, though currently, clinical use of AI is hampered by its complexity and issues concerning data sharing, transparency, and patients safety (He et al , ; Topol, ). Nevertheless, the IDx‐DR software program has been approved by the FDA as the first autonomous diagnostic system for screening of diabetic retinopathy without clinical interpretation (He et al , ).…”

Section: Discussionmentioning

confidence: 99%

“…Therefore, development of software-based support for PoC decision-making is crucial for the success of precision medicine. AI might provide such possibilities in the future, though currently, clinical use of AI is hampered by its complexity and issues concerning data sharing, transparency, and patients safety (He et al, 2019;Topol, 2019 autonomous diagnostic system for screening of diabetic retinopathy without clinical interpretation (He et al, 2019). Unfortunately, at present, datasets tend to be biased toward accessible organ and cell systems and both the quality of publicly available data and the reproducibility of experiments vary, making interpretation difficult.…”

Section: Discussionmentioning

confidence: 99%

4 Ds in health research—working together toward rapid precision medicine

et al. 2019

View full text Add to dashboard Cite

Patient therapy is based mainly on a combination of diagnosis, suitable monitoring or support devices and drug treatment and is usually employed for a pre‐existing disease condition. Therapy remains predominantly symptom‐based, although it is increasingly clear that individual treatment is possible and beneficial. However, reasonable precision medicine can only be realized with the coordinated use of diagnostics, devices and drugs in combination with extensive databases (4Ds), an approach that has not yet found sufficient implementation. The practical combination of 4Ds in health care is progressing, but several obstacles still hamper their extended use in precision medicine.

show abstract

The practical implementation of artificial intelligence technologies in medicine

Cited by 1,365 publications

References 22 publications

Repeatability of radiomics and machine learning for DWI: Short‐term repeatability study of 112 patients with prostate cancer

Repeatability of radiomics and machine learning for DWI: Short‐term repeatability study of 112 patients with prostate cancer

Synthesis of diagnostic quality cancer pathology images

4 Ds in health research—working together toward rapid precision medicine

Contact Info

Product

Resources

About