X-SiT: Inherently Interpretable Surface Vision Transformers for Dementia Diagnosis

Christian Wachinger; Fabian Bongratz; Jaume Gual Ramon; Tom Nuno Wolf

REVIEW 4 major objections 5 minor 35 references

X-SiT: Inherently Interpretable Surface Vision Transformers for Dementia Diagnosis

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper presents X-SiT, an inherently interpretable surface vision transformer that classifies Alzheimer's disease and frontotemporal dementia with accuracy matching non-interpretable models, explaining each prediction through…

desk verdict A genuinely novel spatial-prototype decoder for cortical surfaces, but the 'inherently interpretable' claim needs stronger faithfulness evidence and an honest abstract. read the letter →

arxiv 2506.20267 v1 pith:GVREBGRY submitted 2025-06-25 cs.GR cs.CVcs.LG

Fabian Bongratz , Tom Nuno Wolf , Jaume Gual Ramon , Christian Wachinger This is my paper · ORCID

classification cs.GRcs.CVcs.LG

keywords explainableAIcorticalsurfaceanalysisvisiontransformerprototypicalnetworkcase-basedreasoningAlzheimer'sdiseasefrontotemporaldementiasurface-basedmorphometry

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper introduces X-SiT, a transformer network that classifies cortical surface data into dementia diagnoses while explaining each prediction by pointing to specific brain regions of real training cases. The central claim is that X-SiT is the first inherently interpretable neural network for cortical surfaces that keeps pace with non-interpretable state-of-the-art models: on Alzheimer's disease detection it reaches a balanced accuracy of 80.2% versus 80.5% for the plain surface transformer, and on frontotemporal dementia 79.6% versus 80.6%. The explanations are not post-hoc saliency maps but the network's actual decision rule: a weighted cosine similarity between encoded surface patches and learned prototypes at anatomically corresponding locations. If this holds, clinicians could review why a model made a diagnosis and spot unreliable predictions without giving up accuracy.

What carries the argument

The central mechanism is the prototypical surface patch (PSP) decoder attached to a Surface Vision Transformer encoder. The input cortical surface is assumed pre-registered to a common template and partitioned into a fixed set of non-overlapping triangular patches, so patch index $i$ denotes the same anatomical region across subjects and prototypes. The encoder maps each patch to a latent embedding; a prototypical sequence $\xi=(\xi_1,\dots,\xi_N)$ is obtained by encoding learned prototype patches. For each index the decoder computes the cosine similarity $\cos(x_i, \xi_i)$ between the input embedding $x_i$ and the prototype embedding $\xi_i$, rectifies all embeddings with ReLU so that absent features cannot increase class probability, scales each similarity by a sparse weight $w_i$ (with $\sum_i w_i=1$ and weights below $1/N$ set to zero), and sums to form $P(c|x)=\sum_i w_i\cos(x_i,\xi_i)$. Every five epochs each prototype's latent representation is replaced by the most similar patch from the training set at the same location, keeping prototypes interpretable as real anatomy. This pipeline is what makes the explanation identical to the decision rule.

What would settle it

Register the same set of brains to a common template twice, for example with different template choices or surface-processing settings, and check whether the same patch indices still correspond to the same anatomy; if the cosine similarities, the sparse weights, or the prototypes change substantially under re-registration, the interpretability claim is tied to the registration pipeline rather than to cortex structure. Alternatively, replace the template with an independently built average surface and see whether the learned prototypes still land on the known disease regions, namely temporal and temporoparietal areas for Alzheimer's disease and frontal and temporal areas for frontotemporal dementia.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that case-based reasoning can replace the opaque classification head of a surface vision transformer without a performance penalty. X-SiT's decoder compares each input patch embedding to a learned prototype embedding at the same spatial index, applies ReLU so that only positive evidence counts, and forms the class probability as a sparse weighted sum of cosine similarities. The prototypes are real training patches refreshed during training, so predictions are literally "this region resembles that region of a patient in the training set." In experiments on two dementia tasks, X-SiT's test balanced accuracy (80.2% for AD, 79.6% for FTD) is competitive with the non-interpretable Surface Vision Transformer (80.5% and 80.6%), and its learned prototypes activate in the temporal and temporoparietal regions for Alzheimer's disease and in frontal and temporal regions for frontotemporal dementia, matching published cortical-thinning patterns. The paper further reports that prototype sets are stable across random initializations (76.3% overlap for AD, 71.7% for FTD) and that individual explanations of misclassified cases deviate from the group-level pattern, suggesting the explanations can flag errors.

Load-bearing premise

The whole explanation pipeline depends on the assumption that after registration to a common template, patch index $i$ names the same anatomical location in every subject's cortex and every prototype, so a cosine similarity at that index is a comparison of matching brain regions; residual registration error or folding variability would make the similarities and the resulting explanations anatomically meaningless.

Editorial extensions

If this is right

X-SiT delivers per-prediction explanations that are the model's actual decision rule, not post-hoc attributions, so a clinician can inspect the exact cortical regions that drove a dementia diagnosis.
The learned prototypes double as global disease maps: the same weights and similarities that classify a patient can be averaged across a cohort to show which regions the model finds characteristic of Alzheimer's disease or frontotemporal dementia.
Because prototypes are replaced with real training patches, every explanation can be traced to concrete patient anatomy, making the model's reasoning auditable against clinical knowledge.
The method matches the balanced accuracy of the non-interpretable surface transformer while outperforming the adapted neural-tree interpretable decoder, indicating that interpretability need not be bought with accuracy in this setting.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the anatomical-correspondence assumption is satisfied across larger, multi-site datasets, the same prototype decoder could be applied to differential diagnosis among several dementia subtypes, where the spatial pattern of atrophy is the key discriminator.
The sparse weighting scheme implies the model deliberately ignores below-average regions; a testable extension would be to compare sparse against dense weighting to see whether diffuse but real disease patterns are being discarded for interpretability's sake.
A natural stress test is to feed X-SiT surfaces registered with different templates or from different preprocessing pipelines; if the prototypes and explanations do not remain stable, the interpretability claim would be tied to a specific registration pipeline rather than to the cortex itself.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The core idea is legitimate and worth a look: anchoring each patch's comparison to a prototype at the exact same patch index on a registered cortical surface. That is the right design for anatomy, unlike global prototype search in ProtoPNet or PIP-Net. The paper makes the architectural case cleanly, and the experiments are run honestly: five seeds, multiple baselines, means and standard deviations. X-SiT stays within noise of SiT on both tasks and beats the adapted ViT-NeT, so the decoder is not an accuracy liability.

The soft spots are in proportion. First, the abstract says 'state-of-the-art performance' while Table 2 shows X-SiT slightly below SiT. The text later says 'competitive'—the abstract should say that. Second, the FTD experiment mixes CN subjects from ADNI and NIFD while all FTD patients come from NIFD. That is a site confound; the model may be picking up scanner differences. It needs a site-harmonization step or at least an analysis of site effects. Third, the interpretability evidence is qualitative, and more importantly, nothing links the latent-space cosine similarity to the raw-feature visualizations. The encoder is trained only with binary cross-entropy, so the prototypes could be similar in latent space for reasons invisible in the rendered thickness maps. A clustering or reconstruction regularizer, or a faithfulness experiment, would make 'inherently interpretable' credible. The paper does none of this. The registration assumption is less of a worry; FreeSurfer registration to FsAverage is standard, so the reader's strongest concern is not the load-bearing one.

None of this sinks the idea. It is a smart adaptation, and the qualitative figures are suggestive. But the central claim is overreaching, and the missing code makes reproduction harder.

Who is this for: people building interpretable models on cortical surfaces, and anyone working on prototype-based explainability. It deserves a serious referee, but the revision needs to address the latent-space faithfulness gap, temper the abstract, and deal with the FTD confound.

Referee Report

4 major / 5 minor

Summary. The manuscript introduces X-SiT, an interpretable transformer for cortical surface classification. A Surface Vision Transformer (SiT) encoder produces patch-wise latent embeddings, and a prototypical surface patch (PSP) decoder computes the class probability as a weighted sum of cosine similarities between encoded patches and learned prototypes that are kept in spatial correspondence. Prototypes are periodically replaced with the most similar patches from training data, so the model is claimed to use case-based reasoning on interpretable cortical features. The method is evaluated for Alzheimer's disease and frontotemporal dementia diagnosis on ADNI and NIFD data, with balanced accuracy comparable to prior non-interpretable models, and qualitative group-level and individual-level visualizations are provided as interpretability evidence.

Significance. The architectural idea is timely and potentially valuable: extending prototype-based interpretability to cortical surface data with template correspondence is a natural and clinically relevant direction, and the sparse weighted decision rule is transparent in structure. The authors report competitive accuracy on two clinical tasks using public datasets, and the prototype-stability overlap statistic is a useful addition. However, the paper's core contribution, "inherent interpretability," is not yet established: the visualizations and evaluations do not demonstrate that the latent-space comparisons underlying the decision rule reflect the raw interpretable features shown to the user. The performance claim in the abstract is also stronger than the test results support. Provided these points are addressed with additional analysis, the method would be a solid contribution to the interpretable medical imaging literature.

major comments (4)

[Section 2.2, Eqs. (1)-(2); Section 2.3; Figure 4] The decision rule is defined entirely on latent embeddings xi and ξi, while the interpretability visualizations (Figure 4) show raw cortical thickness, sulcal depth, and curvature. The encoder is trained with binary cross-entropy only, and there is no loss term (e.g., reconstruction, clustering, or similarity-alignment) that forces high cosine similarity in latent space to correspond to high similarity of the visualized raw features. Therefore, the prototypes may be illustrative rather than faithful: two raw patches that appear very different could be close in latent space if that helps classification. To support the central interpretability claim, the authors must connect latent-space similarity to raw-feature similarity, for example by measuring agreement between latent nearest neighbors and raw-feature nearest neighbors on held-out patches, or by adding an explicit alignment or clustering regularizer. Without such evidence, "inherently interpretable" is not established.
[Section 3.4] The interpretability evaluation is exclusively qualitative. Figure 3 shows mean activation maps and two individual examples, and Figure 4 shows selected prototypes. There is no user study, no quantitative explanation metric (e.g., fidelity, consistency, or agreement with expert-defined regions), and no statistical test that the prototypes align with known disease patterns beyond visual inspection. The claim that prototypes "reveal classification errors" rests on anecdotal cases. As a result, the paper does not demonstrate that X-SiT explanations are more faithful or more useful than the attention weights or saliency maps it criticizes in the introduction. The authors should add a quantitative evaluation of explanation quality, or at minimum an inter-rater or expert-agreement study.
[Abstract; Section 3.2; Table 2] The abstract's claim of "state-of-the-art performance" is not supported by Table 2. On both test sets, the non-interpretable SiT achieves equal or higher balanced accuracy (AD: 80.5±0.5 vs 80.2±2.0; FTD: 80.6±1.8 vs 79.6±1.6), and X-SiT is second to SiT in both cases. The differences are within one standard deviation, so a fair statement is "competitive" or "matches accuracy," which is the phrasing used in the introduction and conclusion. The abstract should be revised to avoid an unsupported claim. Additionally, in the FTD task, cognitively normal subjects are pooled from ADNI and NIFD while patients come only from NIFD, so scanner and protocol differences are a potential confound that should be discussed or mitigated.
[Section 2.1; Section 2.2] The interpretability mechanism relies on patch index i denoting the same anatomical region for every subject and every prototype, stated as an assumption in Section 2.1. The paper does not verify this correspondence after FreeSurfer registration to the FsAverage template. Residual registration error or folding variability could make "corresponding prototypes" and the group-level average maps in Figure 3 anatomically misaligned, which would undermine the clinical meaning of the explanations. A demonstration of registration quality (e.g., overlap of cortical labels on a validation population, or within-subject repeat-scan correspondence) would strengthen the central claim. If correspondence is not reliable, the cosine similarities and the resulting explanations are not anatomically meaningful.

minor comments (5)

[Related work and Table 2] The naming is inconsistent: the related work uses "ViT-Net" while the adapted surface model is called "SiT-NeT" in Table 2 and in the text; please use a single consistent name.
[Section 3.2] There is a typo in the first sentence of Section 3.2: "T able 1" should be "Table 1."
[Section 3.4] The prototype robustness overlap (76.3% and 71.7%) is reported without a definition of how the overlap is computed across random initializations; please specify the metric.
[Figure 3] The yellow ellipses indicating hallmark regions are described in the caption but are difficult to identify in a print version of the figure; consider adding explicit region labels to the figure itself.
[Section 3.4] The statement that the temporal lobe received comparatively low weight in the FTD experiments is somewhat confusing in light of the preceding reference to frontal and temporal regions as FTD hallmarks; please clarify whether the model's focus is expected to be incomplete.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity found: X-SiT's predictions are empirical classification outputs on held-out data, and the prototype/interpretability claims are externally evaluated rather than restatements of model inputs.

full rationale

I examined the derivation chain: the surface encoder maps input patches to latent embeddings; the PSP decoder computes cosine similarities between encoded input patches and learned prototypes (Eq. 1); the class probability is a weighted sparse sum of those similarities (Eq. 2). The prototypes are learned end-to-end with binary cross-entropy and periodically replaced by the most similar training patches at each location. No predicted quantity is defined in terms of the quantity it is said to predict: test-set balanced accuracy is computed on held-out subjects, and the alignment of prototype activations with temporal/frontal disease patterns is checked against external neuroimaging references (e.g., [9,30]), not against the model's own training labels. The only self-citation is reference [33] in a related-work sentence listing prior explainable-medical-imaging efforts; it is not load-bearing for the method, the architecture, or the performance claims. The latent-space-to-raw-feature interpretability question raised by the skeptic is a validity concern, not a circularity: nothing in the paper's equations makes 'prototype looks like input' true by construction. Therefore the derivation is self-contained with respect to circularity.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The model's predictions and explanations depend on learned parameters (SiT encoder, prototypes, sparse scaling weights) and on the assumption that template registration makes patch indices anatomically comparable. The paper introduces no new physical or biological entities.

free parameters (4)

Sparse scaling weights w_1,...,w_N = Learned during end-to-end training
These weights directly determine P(c|x) in Eq. 2 and the relative contribution of each cortical region to the explanation; they are fit to the training data.
Prototype latent embeddings ξ_1,...,ξ_N = Learned and periodically replaced by training patches
Prototypes are the reference points for all cosine-similarity explanations; their choice is fit by the training procedure.
SiT encoder network parameters = Learned end-to-end
The encoder defines the latent space in which all similarities are computed; it is not independently validated.
Architectural and training hyperparameters = Tuned on validation, not reported
Patch count, patch size, encoder depth, number of heads, learning rate, and weight decay are chosen by validation performance; without reporting them, reproduction and the performance claim are incompletely specified.

assumptions (4)

domain assumption Cortical surfaces from all subjects are pre-registered to FsAverage so that patch i is the same anatomical region across subjects.
Spatially corresponding prototypes are only meaningful if registration establishes anatomical correspondence; entered in Sections 2.1 and 2.2.
domain assumption Cosine similarity in the ReLU-rectified latent space is a valid proxy for 'looks like' reasoning about disease patterns.
The interpretability claim rests on this operationalization; it is not directly validated against human judgments.
domain assumption FreeSurfer-derived thickness, sulcal depth, and curvature are informative for dementia classification.
The inputs are standard biomarkers; the paper cites literature but does not validate this assumption itself.
domain assumption Pooling cognitively normal subjects from ADNI and NIFD for the FTD task does not introduce a systematic site or scanner confound.
The FTD classifier is trained on CN from two studies with different acquisition protocols; the paper does not harmonize sites or test for scanner effects.

how reviews work

0 comments

Cite this review

Pith. "Pith review of X-SiT: Inherently Interpretable Surface Vision Transformers for Dementia Diagnosis." pith.science (2026). https://pith.science/paper/GVREBGRY

@misc{pith2026250620267,
  author       = {Pith},
  title        = {Pith review of: X-SiT: Inherently Interpretable Surface Vision Transformers for Dementia Diagnosis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GVREBGRY}},
  note         = {Machine review of arXiv:2506.20267}
}

read the original abstract

Interpretable models are crucial for supporting clinical decision-making, driving advances in their development and application for medical images. However, the nature of 3D volumetric data makes it inherently challenging to visualize and interpret intricate and complex structures like the cerebral cortex. Cortical surface renderings, on the other hand, provide a more accessible and understandable 3D representation of brain anatomy, facilitating visualization and interactive exploration. Motivated by this advantage and the widespread use of surface data for studying neurological disorders, we present the eXplainable Surface Vision Transformer (X-SiT). This is the first inherently interpretable neural network that offers human-understandable predictions based on interpretable cortical features. As part of X-SiT, we introduce a prototypical surface patch decoder for classifying surface patch embeddings, incorporating case-based reasoning with spatially corresponding cortical prototypes. The results demonstrate state-of-the-art performance in detecting Alzheimer's disease and frontotemporal dementia while additionally providing informative prototypes that align with known disease patterns and reveal classification errors.

Figures

Figures reproduced from arXiv: 2506.20267 by the authors.

**Figure 1.** Case-based reasoning with X-SiT. X-SiT learns discriminative prototypes of cortical surfaces; patches are colored in blue and surfaces represent cortical thickness. representations compared to 3D volumetric images is that they preserve spatial relations and topological properties of the cortical sheet, crucial for faithful explanations. Moreover, surfaces are well-suited for 3D visualization; they support interacti… view at source ↗

**Figure 2.** Architecture of the eXplainable Surface Vision Transformer (X-SiT). els. We show that the prototypes learned by X-SiT closely match the discriminative disease patterns from the literature, and we demonstrate their benefit for comprehending the model’s predictions both locally and globally. 2 Methods We illustrate the architecture of our X-SiT model in [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** (a) and (b) show mean activation (weighted similarity to prototypes) in red across correct test set classifications of the dementia class. Higher activation implies a higher contribution to the detection of the respective disease. Yellow ellipses indicate the hallmark regions of the respective dementia type: (a) temporal lobe and temporoparietal junction for Alzheimer’s disease and (b) frontal and temporal regions … view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Prototypical disease-specific input features, i.e., prototypes, learned by X-SiT. We also show the mean features for the CN group in the training set for comparison. Gray areas indicate regions that are ignored by the model (wi = 0, cf. Section 2.2). The visualization …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 27 canonical work pages

[1]

In: Shape in Medical Imaging

Azcona, E.A., Besson, P., Wu, Y., Punjabi, A., Martersteck, A., Dravid, A., Par- rish, T.B., Bandt, S.K., Katsaggelos, A.K.: Interpretation of brain morphology in association to alzheimer’s disease dementia classification using graph convolutional networks on triangulated meshes. In: Shape in Medical Imaging. p. 95–107 (2020)

work page 2020
[2]

Cerebral Cortex33(10), 6474–6485 (2023)

de Chastelaine, M., Srokova, S., Hou, M., Kidwai, A., Kafafi, S.S., Racenstein, M.L., Rugg, M.D.: Cortical thickness, gray matter volume, and cognitive perfor- mance: a crosssectional study of the moderating effects of age on their interrela- tionships. Cerebral Cortex33(10), 6474–6485 (2023)

work page 2023
[3]

Advances in neural information processing systems 32 (2019)

Chen, C., Li, O., Tao, D., Barnett, A., Rudin, C., Su, J.K.: This looks like that: deep learning for interpretable image recognition. Advances in neural information processing systems 32 (2019)

work page 2019
[4]

In: Machine Learning in Medical Imaging

Cheng, J., Zhang, X., Zhao, F., Wu, Z., Yuan, X., Gilmore, J.H., Wang, L., Lin, W., Li, G.: Spherical transformer on cortical surfaces. In: Machine Learning in Medical Imaging. p. 406–415 (2022)

work page 2022
[5]

Molecular Psychiatry28(10), 4084–4097 (2023)

Chouliaras, L., O’Brien, J.T.: The use of neuroimaging techniques in the early and differential diagnosis of dementia. Molecular Psychiatry28(10), 4084–4097 (2023)

work page 2023
[6]

In: Medical Imaging with Deep Learn- ing

Dahan, S., Fawaz, A., Williams, L.Z.J., Yang, C., Coalson, T.S., Glasser, M.F., Ed- wards, A.D., Rueckert, D., Robinson, E.C.: Surface vision transformers: Attention- based modelling applied to cortical analysis. In: Medical Imaging with Deep Learn- ing. Proceedings of Machine Learning Research, vol. 172, pp. 282–303 (2022)

work page 2022
[7]

NeuroImage 9(2), 179–194 (1999)

Dale, A.M., Fischl, B., Sereno, M.I.: Cortical surface-based analysis i. NeuroImage 9(2), 179–194 (1999)

work page 1999
[8]

In: International Conference on Learn- ing Representations (2021)

Dosovitskiy, A., Beyer, L., Kolesnikov, A., et al.: An image is worth 16x16 words: Transformers for image recognition at scale. In: International Conference on Learn- ing Representations (2021)

work page 2021

Show all 35 references

[9]

Brain130(4), 1159–1166 (2006)

Du, A.T., Schuff, N., Kramer, J.H., Rosen, H.J., Gorno-Tempini, M.L., Rankin, K., Miller, B.L., Weiner, M.W.: Different regional patterns of cortical thinning in alzheimer’s disease and frontotemporal dementia. Brain130(4), 1159–1166 (2006)

2006
[10]

NeuroImage62(2), 774–781 (2012)

Fischl, B.: FreeSurfer. NeuroImage62(2), 774–781 (2012)

2012
[11]

NeuroImage 9(2), 195–207 (1999) 10 F

Fischl, B., Sereno, M.I., Dale, A.M.: Cortical surface-based analysis ii. NeuroImage 9(2), 195–207 (1999) 10 F. Bongratz, T.N. Wolf, et al

1999
[12]

Gallée, L., Beer, M., Götz, M.: Interpretable medical image classification using pro- totypelearningandprivilegedinformation.In:InternationalConferenceonMedical Image Computing and Computer-Assisted Intervention. pp. 435–445 (2023)

2023
[13]

In: 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)

Gong, S., Chen, L., Bronstein, M., Zafeiriou, S.: Spiralnet++: A fast and highly efficient mesh convolution operator. In: 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW). p. 4141–4148 (2019)

2019
[14]

IEEE Transactions on Pattern Analysis and Machine Intelligence44(2), 864–876 (2022)

Gopinath, K., Desrosiers, C., Lombaert, H.: Learnable pooling in graph convolu- tional networks for brain surface analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence44(2), 864–876 (2022)

2022
[15]

Advances in Neural Information Processing Systems33, 4211–4222 (2020)

Jeyakumar, J.V., Noor, J., Cheng, Y.H., Garcia, L., Srivastava, M.: How can i ex- plain this to you? an empirical study of deep neural network explanation methods. Advances in Neural Information Processing Systems33, 4211–4222 (2020)

2020
[16]

Kashefi, R., Barekatain, L., Sabokrou, M., Aghaeipoor, F.: Explainability of vision transformers: A comprehensive review and new perspectives. arxiv. arXiv preprint arXiv:2311.06786 (2023)

2023 arXiv
[17]

In: International Conference on Machine Learning

Kim,S.,Nam,J.,Ko,B.C.:ViT-NeT:Interpretablevisiontransformerswithneural tree decoder. In: International Conference on Machine Learning. Proceedings of Machine Learning Research, vol. 162, pp. 11162–11172 (2022)

2022
[18]

help me help the ai

Kim, S.S., Watkins, E.A., Russakovsky, O., Fong, R., Monroy-Hernández, A.: "help me help the ai": Understanding how explainability can support human-ai interac- tion. In: Conference on Human Factors in Computing Systems. pp. 1–17 (2023)

2023
[19]

In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

Komorowski, P., Baniecki, H., Biecek, P.: Towards evaluating explanations of vision transformers for medical imaging. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 3726–3732 (2023)

2023
[20]

In: International Conference on Learning Representations (2019)

Loshchilov, I., Hutter, F.: Decoupled weight decay regularization. In: International Conference on Learning Representations (2019)

2019
[21]

In: Canadian AI (2021)

Mohammadjafari, S., Cevik, M., Thanabalasingam, M., Basar, A.: Using protopnet for interpretable alzheimer’s disease classification. In: Canadian AI (2021)

2021
[22]

npj Digital Medicine7(1) (2024)

Pahud de Mortanges, A., Luo, H., Shu, S.Z., Kamath, A., Suter, Y., Shelan, M., Pöllinger, A., Reyes, M.: Orchestrating explainable artificial intelligence for multi- modal and longitudinal data in medical imaging. npj Digital Medicine7(1) (2024)

2024
[23]

Imaging Neuroscience2, 1–37 (2024)

Munroe, L., da Silva, M., Heidari, F., Grigorescu, I., Dahan, S., Robinson, E.C., Deprez, M., So, P.W.: Applications of interpretable deep learning in neuroimaging: A comprehensive review. Imaging Neuroscience2, 1–37 (2024)

2024
[24]

Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (2023)

Nauta, M., Schlötterer, J., van Keulen, M., Seifert, C.: Pip-net: Patch-based intuitive prototypes for interpretable image classification. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (2023)

2023
[25]

Advances in Neural Informa- tion Processing Systems34, 26422–26436 (2021)

Nguyen, G., Kim, D., Nguyen, A.: The effectiveness of feature attribution methods and its correlation with automatic evaluation scores. Advances in Neural Informa- tion Processing Systems34, 26422–26436 (2021)

2021
[26]

IEEE Transactions on Medical Imaging19(2), 143–150 (2000)

Nyul, L., Udupa, J., Zhang, X.: New variants of a method of mri scale standard- ization. IEEE Transactions on Medical Imaging19(2), 143–150 (2000)

2000
[27]

Nature 618(7965), 566–574 (2023)

Pang, J.C., Aquino, K.M., Oldehinkel, M., Robinson, P.A., Fulcher, B.D., Break- spear, M., Fornito, A.: Geometric constraints on human brain function. Nature 618(7965), 566–574 (2023)

2023
[28]

Nature machine intelligence1(5), 206–215 (2019)

Rudin, C.: Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nature machine intelligence1(5), 206–215 (2019)

2019
[29]

Statistic Sur- veys 16, 1–85 (2022) X-SiT 11

Rudin, C., Chen, C., Chen, Z., Huang, H., Semenova, L., Zhong, C.: Interpretable machine learning: Fundamental principles and 10 grand challenges. Statistic Sur- veys 16, 1–85 (2022) X-SiT 11

2022
[30]

Brain 129(11), 2885–2893 (2006)

Singh, V., Chertkow, H., Lerch, J.P., Evans, A.C., Dorr, A.E., Kabani, N.J.: Spatial patterns of cortical thinning in mild cognitive impairment and alzheimer’s disease. Brain 129(11), 2885–2893 (2006)

2006
[31]

Brain 145(11), 3859–3871 (2022)

Spitzer, H., Ripart, M., Whitaker, K., et al.: Interpretable surface-based detection of focal cortical dysplasias: a multi-centre epilepsy lesion detection study. Brain 145(11), 3859–3871 (2022)

2022
[32]

Journal of the Royal Statistical Society Series B: Statistical Methodology58(1), 267–288 (1996)

Tibshirani, R.: Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society Series B: Statistical Methodology58(1), 267–288 (1996)

1996
[33]

In: International Conference on Information Processing in Medical Imaging

Wolf, T.N., Pölsterl, S., Wachinger, C.: Don’t panic: Prototypical additive neu- ral network for interpretable classification of alzheimer’s disease. In: International Conference on Information Processing in Medical Imaging. pp. 82–94 (2023)

2023
[34]

Information Fusion77, 29–52 (2022)

Yang, G., Ye, Q., Xia, J.: Unbox the black-box for the medical explainable ai via multi-modal and multi-centre data fusion: A mini-review, two showcases and beyond. Information Fusion77, 29–52 (2022)

2022
[35]

Zhao, F., Xia, S., Wu, Z., Duan, D., Wang, L., Lin, W., Gilmore, J.H., Shen, D., Li, G.: Spherical U-Net on Cortical Surfaces: Methods and Applications, p. 855–866. Springer International Publishing (2019)

2019

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

In: Shape in Medical Imaging

Azcona, E.A., Besson, P., Wu, Y., Punjabi, A., Martersteck, A., Dravid, A., Par- rish, T.B., Bandt, S.K., Katsaggelos, A.K.: Interpretation of brain morphology in association to alzheimer’s disease dementia classification using graph convolutional networks on triangulated meshes. In: Shape in Medical Imaging. p. 95–107 (2020)

work page 2020

[2] [2]

Cerebral Cortex33(10), 6474–6485 (2023)

de Chastelaine, M., Srokova, S., Hou, M., Kidwai, A., Kafafi, S.S., Racenstein, M.L., Rugg, M.D.: Cortical thickness, gray matter volume, and cognitive perfor- mance: a crosssectional study of the moderating effects of age on their interrela- tionships. Cerebral Cortex33(10), 6474–6485 (2023)

work page 2023

[3] [3]

Advances in neural information processing systems 32 (2019)

Chen, C., Li, O., Tao, D., Barnett, A., Rudin, C., Su, J.K.: This looks like that: deep learning for interpretable image recognition. Advances in neural information processing systems 32 (2019)

work page 2019

[4] [4]

In: Machine Learning in Medical Imaging

Cheng, J., Zhang, X., Zhao, F., Wu, Z., Yuan, X., Gilmore, J.H., Wang, L., Lin, W., Li, G.: Spherical transformer on cortical surfaces. In: Machine Learning in Medical Imaging. p. 406–415 (2022)

work page 2022

[5] [5]

Molecular Psychiatry28(10), 4084–4097 (2023)

Chouliaras, L., O’Brien, J.T.: The use of neuroimaging techniques in the early and differential diagnosis of dementia. Molecular Psychiatry28(10), 4084–4097 (2023)

work page 2023

[6] [6]

In: Medical Imaging with Deep Learn- ing

Dahan, S., Fawaz, A., Williams, L.Z.J., Yang, C., Coalson, T.S., Glasser, M.F., Ed- wards, A.D., Rueckert, D., Robinson, E.C.: Surface vision transformers: Attention- based modelling applied to cortical analysis. In: Medical Imaging with Deep Learn- ing. Proceedings of Machine Learning Research, vol. 172, pp. 282–303 (2022)

work page 2022

[7] [7]

NeuroImage 9(2), 179–194 (1999)

Dale, A.M., Fischl, B., Sereno, M.I.: Cortical surface-based analysis i. NeuroImage 9(2), 179–194 (1999)

work page 1999

[8] [8]

In: International Conference on Learn- ing Representations (2021)

Dosovitskiy, A., Beyer, L., Kolesnikov, A., et al.: An image is worth 16x16 words: Transformers for image recognition at scale. In: International Conference on Learn- ing Representations (2021)

work page 2021