Seeing the random forest through the decision trees. Supporting learning health systems from histopathology with machine learning models: Challenges and opportunities

Gonzalez, Ricardo; Saha, Ashirbani; Campbell, Clinton J.V.; Nejat, Peyman; Lokker, Cynthia; Norgan, Andrew P.

doi:10.1016/j.jpi.2023.100347

Cited by 2 publications

(3 citation statements)

References 213 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These thresholds are chosen for each iteration at the end of training, and so will change over time. (7) Leave sections with f < f thresh in the active set. Sections with f ≥ f thresh are evaluated by the expert to accept or reject for inclusion in the training set.…”

Section: Active Deep Learning (Adl)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…Deep learning and convolutional neural networks have proven to be highly effective and useful tools in medical image analysis, including MRI, CT, PET, and histopathology images [1][2][3][4][5][6][7][8]. Even so, expert labeling of medical image datasets to generate ground truth for training deep learning algorithms is still an ongoing challenge to the adoption of artificial intelligence in medical practice [7]. In an ideal world, all possible medical images would be curated, labeled, collected, and available at any time for anyone to train universally robust machine learning algorithms, and significant efforts have been devoted to generating medical image datasets with ground truth for research communities [9].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Synergizing Deep Learning-Enabled Preprocessing and Human–AI Integration for Efficient Automatic Ground Truth Generation

Collazo,

Vargas,

Cara

et al. 2024

Bioengineering

View full text Add to dashboard Cite

The progress of incorporating deep learning in the field of medical image interpretation has been greatly hindered due to the tremendous cost and time associated with generating ground truth for supervised machine learning, alongside concerns about the inconsistent quality of images acquired. Active learning offers a potential solution to these problems of expanding dataset ground truth by algorithmically choosing the most informative samples for ground truth labeling. Still, this effort incurs the costs of human labeling, which needs minimization. Furthermore, automatic labeling approaches employing active learning often exhibit overfitting tendencies while selecting samples closely aligned with the training set distribution and excluding out-of-distribution samples, which could potentially improve the model’s effectiveness. We propose that the majority of out-of-distribution instances can be attributed to inconsistent cross images. Since the FDA approved the first whole-slide image system for medical diagnosis in 2017, whole-slide images have provided enriched critical information to advance the field of automated histopathology. Here, we exemplify the benefits of a novel deep learning strategy that utilizes high-resolution whole-slide microscopic images. We quantitatively assess and visually highlight the inconsistencies within the whole-slide image dataset employed in this study. Accordingly, we introduce a deep learning-based preprocessing algorithm designed to normalize unknown samples to the training set distribution, effectively mitigating the overfitting issue. Consequently, our approach significantly increases the amount of automatic region-of-interest ground truth labeling on high-resolution whole-slide images using active deep learning. We accept 92% of the automatic labels generated for our unlabeled data cohort, expanding the labeled dataset by 845%. Additionally, we demonstrate expert time savings of 96% relative to manual expert ground-truth labeling.

show abstract

Section: Active Deep Learning (Adl)mentioning

confidence: 99%