Explainable Deep CNNs for MRI-Based Diagnosis of Alzheimer’s Disease

Nigri, Eduardo; Ziviani, Nívio; Cappabianco, Fábio A. M.; Antunes, Augusto; Veloso, Adriano

doi:10.1109/ijcnn48605.2020.9206837

Cited by 36 publications

(16 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Future research could adapt our framework to assess the transportability of prediction models when intervening on endogenous variables and causal relationships, because real-world populations likely do not only differ by exogenous variables. Recent ML models predicted AD from medical images [7,50], which requires new approaches to identify the causal structure in complex data [51].…”

Section: Discussionmentioning

confidence: 99%

Assessing the transportability of clinical prediction models for cognitive impairment using causal models

Fehr

Piccininni

Kurth

et al. 2022

Preprint

View full text Add to dashboard Cite

BACKGROUND: Machine learning promises to support the diagnosis of dementia and Alzheimer's Disease, but may not perform well in new settings. We present a framework to assess the transportability of models predicting cognitive impairment in external settings with different demographics. METHODS: We mapped and quantified relationships between variables associated with cognitive impairment using causal graphs, structural equation models, and data from the ADNI study. These estimates generated datasets for training and validating prediction models. We measured transportability to external settings with interventions on age, APOE e4, and sex, using calibration metric differences. RESULTS: Models predicting with causes of the outcome were 1.3-12.8 times more transportable than those predicting with consequences. Logistic and lasso models had better calibration in internal validation settings than random forest and boosted models. DISCUSSION: Applying a framework considering causal relationships is crucial to assess transportability. Future research could investigate more interventions and methods to quantify causal relationships in risk prediction.

show abstract

Section: Discussionmentioning

confidence: 99%

Assessing the transportability of clinical prediction models for cognitive impairment using causal models

Fehr

Piccininni

Kurth

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The standard perturbation method has been widely used in the study of Alzheimer's disease [32,48,45,37] and related symptoms (amyloid-β pathology) [49]. However, most of the time, authors do not train their model with perturbed images.…”

Section: Perturbation Methods Applied To Neuroimagingmentioning

confidence: 99%

“…More advanced perturbation based methods have also been used in the literature. Nigri et al [45] compared a classical perturbation method to a swap test. The swap test replaces the classical perturbation step by a swapping step where patches are exchanged between the input brain image and a reference image chosen according to the model prediction.…”

Section: Advanced Perturbation Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Interpretability of Machine Learning Methods Applied to Neuroimaging

Thibeau–Sutre¹,

Collin²,

Burgos³

et al. 2022

Preprint

View full text Add to dashboard Cite

Deep learning methods have become very popular for the processing of natural images, and were then successfully adapted to the neuroimaging field. As these methods are non-transparent, interpretability methods are needed to validate them and ensure their reliability. Indeed, it has been shown that deep learning models may obtain high performance even when using irrelevant features, by exploiting biases in the training set. Such undesirable situations can potentially be detected by using interpretability methods. Recently, many methods have been proposed to interpret neural networks. However, this domain is not mature yet. Machine learning users face two major issues when aiming to interpret their models: which method to choose, and how to assess its reliability? Here, we aim at providing answers to these questions by presenting the most common interpretability methods and metrics developed to assess their reliability, as well as their applications and benchmarks in the neuroimaging context. Note that this is not an exhaustive survey: we aimed to focus on the studies which we found to be the most representative and relevant.

show abstract

“…For the second category, there were three strategies to visualize features for CNN. 1) By editing an input image and observing its effect on the prediction results, the occluded regions which had a significant impact on prediction can be visualized [30]. For instance, Zeiler and Fergus [31] proposed an occlusion-based method to visualize the activity within CNN.…”

Section: Related Workmentioning

confidence: 99%

Morphological feature visualization of Alzheimer's disease via Multidirectional Perception GAN

Yu¹,

Lei²,

Shen³

et al. 2021

Preprint

View full text Add to dashboard Cite

The diagnosis of early stages of Alzheimer's disease (AD) is essential for timely treatment to slow further deterioration. Visualizing the morphological features for the early stages of AD is of great clinical value. In this work, a novel Multidirectional Perception Generative Adversarial Network (MP-GAN) is proposed to visualize the morphological features indicating the severity of AD for patients of different stages. Specifically, by introducing a novel multidirectional mapping mechanism into the model, the proposed MP-GAN can capture the salient global features efficiently. Thus, by utilizing the classdiscriminative map from the generator, the proposed model can clearly delineate the subtle lesions via MR image transformations between the source domain and the pre-defined target domain. Besides, by integrating the adversarial loss, classification loss, cycle consistency loss and L1 penalty, a single generator in MP-GAN can learn the class-discriminative maps for multipleclasses. Extensive experimental results on Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset demonstrate that MP-GAN achieves superior performance compared with the existing methods. The lesions visualized by MP-GAN are also consistent with what clinicians observe.

show abstract

Explainable Deep CNNs for MRI-Based Diagnosis of Alzheimer’s Disease

Cited by 36 publications

References 31 publications

Assessing the transportability of clinical prediction models for cognitive impairment using causal models

Assessing the transportability of clinical prediction models for cognitive impairment using causal models

Interpretability of Machine Learning Methods Applied to Neuroimaging

Morphological feature visualization of Alzheimer's disease via Multidirectional Perception GAN

Contact Info

Product

Resources

About