arxiv: 2602.07131 · v2 · submitted 2026-02-06 · 📡 eess.SP · q-bio.NC

Recognition: 2 theorem links

· Lean Theorem

Behavior Score Prediction in Resting-State Functional MRI by Deep State Space Modeling

Javier Salazar Cavazos , Maximillian Egan , Krisanne Litinas , Benjamin Hampstead , Scott Peltier

Authors on Pith no claims yet

Pith reviewed 2026-05-16 05:55 UTC · model grok-4.3

classification 📡 eess.SP q-bio.NC

keywords Alzheimer's diseaseresting-state fMRIstate space modelsbehavior score predictionBOLD time seriescognitive impairmentbrain regions

0 comments

The pith

A deep state space model predicts Alzheimer's behavior scores from fMRI by learning sparse brain regions from BOLD time series.

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

The paper shows that deep state space models can forecast behavior scores measuring language, memory, and cognitive skills in Alzheimer's by working directly with the full BOLD time series from resting-state fMRI. Prior approaches collapse the data into static functional connectivity matrices and lose the temporal patterns of brain activity. The new framework learns a small set of key brain regions whose dynamic signals better predict impairment. Tests on Michigan Alzheimer's data indicate higher accuracy than connectivity baselines and point to specific regions tied to early pathology.

Core claim

Our deep state space modeling framework directly leverages the blood-oxygenation-level-dependent time series to learn a sparse collection of brain regions to predict behavior scores. Our model extracts temporal features that encapsulate nuanced patterns of intrinsic brain activity, thereby enhancing predictive performance compared to traditional connectivity methods. We identify specific brain regions that are most predictive of cognitive impairment through experiments on data provided by the Michigan Alzheimer's Disease Research Center, providing new insights into the neural substrates of early Alzheimer's pathology.

What carries the argument

Deep state space model that processes full BOLD time series to extract temporal features and select a sparse set of predictive brain regions for behavior score forecasting.

If this is right

Retaining the full temporal structure of BOLD signals improves prediction of cognitive scores compared with averaged connectivity matrices.
The learned sparse brain regions supply concrete targets for studying the neural basis of early Alzheimer's changes.
The framework supports risk monitoring strategies that rely on imaging time series rather than static summaries.
Temporal feature extraction can highlight dynamic activity patterns missed by traditional methods.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

If the sparse regions hold up across multiple sites, they could become practical imaging markers for tracking progression.
The same time-series approach might apply to predicting outcomes in other disorders where brain dynamics shift over time.
Longitudinal fMRI collections could let the model forecast score changes rather than single-time-point values.
Combining the state-space output with clinical variables might further reduce prediction error on new patients.

Load-bearing premise

The model can reliably pick a small number of brain regions from the fMRI signals that predict behavior scores without overfitting or needing much more data to check beyond one dataset.

What would settle it

Running the model on a new independent set of resting-state fMRI scans and finding no gain in prediction accuracy over connectivity methods or no replication of the same predictive regions.

read the original abstract

Early clinical assessment of Alzheimer's disease relies on behavior scores that measure a subject's language, memory, and cognitive skills. On the medical imaging side, functional magnetic resonance imaging has provided invaluable insights into the neural pathways underlying Alzheimer's disease. While prior studies have used resting-state functional MRI by extracting functional connectivity matrices, these approaches neglect the temporal dynamics inherent in functional data. In this work, we present a deep state space modeling framework that directly leverages the blood-oxygenation-level-dependent time series to learn a sparse collection of brain regions to predict behavior scores. Our model extracts temporal features that encapsulate nuanced patterns of intrinsic brain activity, thereby enhancing predictive performance compared to traditional connectivity methods. We identify specific brain regions that are most predictive of cognitive impairment through experiments on data provided by the Michigan Alzheimer's Disease Research Center, providing new insights into the neural substrates of early Alzheimer's pathology. These findings have important implications for the possible development of risk monitoring and intervention strategies in Alzheimer's disease.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper applies deep state space models to raw BOLD time series for Alzheimer's behavior score prediction but the abstract supplies no numbers or validation details to support the claimed gains.

read the letter

The core move here is shifting from functional connectivity matrices to direct modeling of BOLD time series with deep state space models, then using that to pick a sparse set of predictive regions for behavior scores on the Michigan ADRC cohort. That framing addresses a real gap in the literature they cite, where temporal structure gets collapsed away early. The paper does a clean job stating the clinical motivation around early assessment and laying out why intrinsic dynamics might matter for cognitive impairment signals. If the full methods section shows how the latent states are learned and how sparsity is enforced, that part could be useful to people already working on state-space approaches in neuroimaging. The results section apparently identifies specific regions tied to the predictions, which at least gives a concrete output to inspect. The main weakness is that the abstract reports no quantitative metrics, no cross-validation scheme, no baseline numbers, and no error bars, so the performance claim stays untestable from what is visible. The single-site dataset is also a clear limit; rs-fMRI signals carry scanner and population effects that are hard to rule out without held-out cohorts, and the stress-test note is right that the learned sparsity pattern could be overfit to this particular sample. Minor issues include the usual need for more detail on latent dimensionality and training stability, but those are fixable in revision. This is aimed at neuroimaging groups doing Alzheimer's prediction or temporal modeling of fMRI; a reader already familiar with SSMs or looking for alternatives to connectivity matrices would get the most out of it. The work is coherent enough on its own terms to deserve referee time, even if the evidence needs substantial strengthening.

Referee Report

2 major / 0 minor

Summary. The paper proposes a deep state space modeling framework that processes resting-state fMRI BOLD time series directly to learn a sparse set of predictive brain regions for behavior-score prediction in early Alzheimer's disease. It claims that the extracted temporal features capture nuanced intrinsic brain activity patterns and yield better predictive performance than traditional functional connectivity matrices, with experiments on the Michigan ADRC dataset identifying specific regions linked to cognitive impairment.

Significance. If the claimed performance gains and region identifications hold under rigorous validation, the work would advance rs-fMRI analysis by moving beyond static connectivity to explicit temporal modeling, potentially improving early AD risk assessment and identifying actionable neural targets for monitoring or intervention.

major comments (2)

[Abstract] Abstract: the central claim of 'enhancing predictive performance compared to traditional connectivity methods' is asserted without any quantitative metrics, error bars, cross-validation scheme, baseline specifications, or statistical tests, leaving the performance advantage unsupported by visible evidence in the provided text.
[Abstract] Abstract / Experiments description: all reported results, including the learned sparse region set and performance gains, derive from the single Michigan ADRC cohort. Given well-documented scanner-, site-, and population-specific variance in rs-fMRI BOLD signals, the sparsity pattern and claimed superiority cannot be distinguished from dataset-specific overfitting without held-out external cohorts or multi-site replication.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for the constructive feedback. We address each major comment below, indicating planned revisions where appropriate.

read point-by-point responses

Referee: [Abstract] Abstract: the central claim of 'enhancing predictive performance compared to traditional connectivity methods' is asserted without any quantitative metrics, error bars, cross-validation scheme, baseline specifications, or statistical tests, leaving the performance advantage unsupported by visible evidence in the provided text.

Authors: We agree that the abstract should include quantitative support for the performance claims. The full manuscript reports results from 5-fold cross-validation on the Michigan ADRC dataset, with explicit baselines (functional connectivity matrices), metrics (e.g., MSE and correlation), error bars, and statistical tests. We will revise the abstract to incorporate these key quantitative elements while maintaining brevity. revision: yes
Referee: [Abstract] Abstract / Experiments description: all reported results, including the learned sparse region set and performance gains, derive from the single Michigan ADRC cohort. Given well-documented scanner-, site-, and population-specific variance in rs-fMRI BOLD signals, the sparsity pattern and claimed superiority cannot be distinguished from dataset-specific overfitting without held-out external cohorts or multi-site replication.

Authors: We acknowledge the single-cohort limitation and the risk of dataset-specific effects in rs-fMRI. The Michigan ADRC data were collected under standardized protocols, and we used internal cross-validation plus sparsity regularization to mitigate overfitting. We will expand the discussion to explicitly note this limitation and call for future multi-site validation. However, we do not have access to additional external cohorts for the current study. revision: partial

standing simulated objections not resolved

Absence of external validation on held-out multi-site cohorts, preventing confirmation that the identified sparse regions and performance gains generalize beyond the Michigan ADRC dataset.

Circularity Check

0 steps flagged

No significant circularity in the derivation chain

full rationale

The paper presents an empirical deep state-space model trained directly on BOLD time series to predict behavior scores and identify sparse predictive regions. No equations, derivations, or load-bearing steps reduce any claimed prediction to a fitted parameter or input by construction. The framework is data-driven with external validation against connectivity baselines on the Michigan ADRC cohort; no self-definitional loops, ansatz smuggling, or uniqueness theorems imported from prior self-citations appear in the modeling chain. This is a standard ML application without circular reduction.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The framework rests on standard neuroimaging assumptions about BOLD signals reflecting neural dynamics and on typical deep learning choices for state space models; no new entities are introduced.

free parameters (1)

latent state dimensionality
Standard hyperparameter in state space models, chosen or fitted to capture brain dynamics.

axioms (1)

domain assumption BOLD time series contain learnable temporal patterns predictive of behavior scores
Core premise of the modeling approach stated in the abstract.

pith-pipeline@v0.9.0 · 5474 in / 1012 out tokens · 54358 ms · 2026-05-16T05:55:37.694752+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

Our model extracts temporal features that encapsulate nuanced patterns of intrinsic brain activity... NeuroMamba backbone... Mamba++ Layer... min θ ½∥s−gθ(fθ(Xi))∥²₂ + λ∥fθ(Xi)∥₁
IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

deep state space modeling framework that directly leverages the blood-oxygenation-level-dependent time series

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

87 extracted references · 87 canonical work pages · 2 internal anchors

[1]

Clinical applications and future directions of functional mri,

D. A. Orringer, D. R. Vago, and A. J. Golby, “Clinical applications and future directions of functional mri,” inSeminars in neurology, vol. 32, no. 04. Thieme Medical Publishers, 2012, pp. 466–475

work page 2012
[2]

Opportunities and limitations of intrinsic functional connectivity mri,

R. L. Buckner, F. M. Krienen, and B. T. Yeo, “Opportunities and limitations of intrinsic functional connectivity mri,”Nature neuroscience, vol. 16, no. 7, pp. 832–837, 2013

work page 2013
[3]

Functional connectivity in the motor cortex of resting human brain using echo- planar mri,

B. Biswal, F. Zerrin Yetkin, V . M. Haughton, and J. S. Hyde, “Functional connectivity in the motor cortex of resting human brain using echo- planar mri,”Magnetic resonance in medicine, vol. 34, no. 4, pp. 537– 541, 1995

work page 1995
[4]

Biophysical modulations of functional connectivity,

S. J. Peltier and Y . Shah, “Biophysical modulations of functional connectivity,”Brain connectivity, vol. 1, no. 4, pp. 267–277, 2011

work page 2011
[5]

The functional brain organization of an individual allows pre- diction of measures of social abilities transdiagnostically in autism and attention-deficit/hyperactivity disorder,

E. M. Lake, E. S. Finn, S. M. Noble, T. Vanderwal, X. Shen, M. D. Rosenberg, M. N. Spann, M. M. Chun, D. Scheinost, and R. T. Con- stable, “The functional brain organization of an individual allows pre- diction of measures of social abilities transdiagnostically in autism and attention-deficit/hyperactivity disorder,”Biological psychiatry, vol. 86, no. 4,...

work page 2019
[7]

Functional connectivity in the resting brain: a network analysis of the default mode hypothesis,

M. D. Greicius, B. Krasnow, A. L. Reiss, and V . Menon, “Functional connectivity in the resting brain: a network analysis of the default mode hypothesis,”Proceedings of the national academy of sciences, vol. 100, no. 1, pp. 253–258, 2003

work page 2003
[8]

A default mode of brain function,

M. E. Raichle, A. M. MacLeod, A. Z. Snyder, W. J. Powers, D. A. Gusnard, and G. L. Shulman, “A default mode of brain function,” Proceedings of the national academy of sciences, vol. 98, no. 2, pp. 676–682, 2001

work page 2001
[9]

Nia-aa research framework: toward a biological definition of alzheimer’s disease,

C. R. Jack Jr, D. A. Bennett, K. Blennow, M. C. Carrillo, B. Dunn, S. B. Haeberlein, D. M. Holtzman, W. Jagust, F. Jessen, J. Karlawish et al., “Nia-aa research framework: toward a biological definition of alzheimer’s disease,”Alzheimer’s & dementia, vol. 14, no. 4, pp. 535– 562, 2018

work page 2018
[10]

Struc- tural brain abnormalities common to posttraumatic stress disorder and depression,

M. C. Kroes, M. D. Rugg, M. G. Whalley, and C. R. Brewin, “Struc- tural brain abnormalities common to posttraumatic stress disorder and depression,”Journal of Psychiatry and Neuroscience, vol. 36, no. 4, pp. 256–265, 2011

work page 2011
[11]

The montreal cognitive assessment, moca: a brief screening tool for mild cognitive impairment,

Z. S. Nasreddine, N. A. Phillips, V . B ´edirian, S. Charbonneau, V . White- head, I. Collin, J. L. Cummings, and H. Chertkow, “The montreal cognitive assessment, moca: a brief screening tool for mild cognitive impairment,”Journal of the American Geriatrics Society, vol. 53, no. 4, pp. 695–699, 2005

work page 2005
[12]

The use of random forests to identify brain regions on amyloid and fdg pet associated with moca score,

K. Zukotynski, V . Gaudet, P. H. Kuo, S. Adamo, M. Goubran, C. J. Scott, C. Bocti, M. Borrie, H. Chertkow, R. Frayneet al., “The use of random forests to identify brain regions on amyloid and fdg pet associated with moca score,”Clinical nuclear medicine, vol. 45, no. 6, pp. 427–433, 2020

work page 2020
[13]

Individualized prediction of cognitive test scores from functional brain connectome in patients with first-episode late-life depression,

C. Liu, L. Li, D. Zhu, S. Lin, L. Ren, W. Zhen, W. Tan, L. Wang, L. Tian, Q. Wanget al., “Individualized prediction of cognitive test scores from functional brain connectome in patients with first-episode late-life depression,”Journal of Affective Disorders, vol. 352, pp. 32–42, 2024

work page 2024
[14]

Using connectome-based predictive modeling to predict individual behavior from brain connectivity,

X. Shen, E. S. Finn, D. Scheinost, M. D. Rosenberg, M. M. Chun, X. Pa- pademetris, and R. T. Constable, “Using connectome-based predictive modeling to predict individual behavior from brain connectivity,”nature protocols, vol. 12, no. 3, pp. 506–518, 2017

work page 2017
[15]

Resting-state functional connectivity predicts cognitive impair- ment related to alzheimer’s disease,

Q. Lin, M. D. Rosenberg, K. Yoo, T. W. Hsu, T. P. O’Connell, and M. M. Chun, “Resting-state functional connectivity predicts cognitive impair- ment related to alzheimer’s disease,”Frontiers in aging neuroscience, vol. 10, p. 94, 2018

work page 2018
[16]

A new rating scale for alzheimer’s disease

W. G. Rosen, R. C. Mohs, and K. L. Davis, “A new rating scale for alzheimer’s disease.”The American journal of psychiatry, vol. 141, no. 11, pp. 1356–1364, 1984

work page 1984
[17]

Predictive analysis and deep learning of functional mri in alzheimer’s disease,

M. Karker, “Predictive analysis and deep learning of functional mri in alzheimer’s disease,” Ph.D. dissertation, University of Michigan, 2022

work page 2022
[18]

Synergistic effects of cognitive training and high-definition transcranial direct current stimulation across the dementia spectrum,

B. M. Hampstead, A. Iordan, A. Rahman-Filipiak, R. Ploutz-Snyder, and P. Pruitt, “Synergistic effects of cognitive training and high-definition transcranial direct current stimulation across the dementia spectrum,” Brain Stimulation: Basic, Translational, and Clinical Research in Neu- romodulation, vol. 18, no. 1, p. 350, 2025

work page 2025
[19]

Ultrasound brain stimulation technologies for targeted therapeutics,

Y . Jo, S. Kim, J. Jeong, and H. J. Lee, “Ultrasound brain stimulation technologies for targeted therapeutics,”Nature Electronics, vol. 8, no. 8, pp. 647–662, 2025

work page 2025
[20]

The alzheimer’s disease centers’ uniform data set (uds): the neuropsychologic test battery,

S. Weintraub, D. Salmon, N. Mercaldo, S. Ferris, N. R. Graff-Radford, H. Chui, J. Cummings, C. DeCarli, N. L. Foster, D. Galaskoet al., “The alzheimer’s disease centers’ uniform data set (uds): the neuropsychologic test battery,”Alzheimer Disease & Associated Disorders, vol. 23, no. 2, pp. 91–101, 2009

work page 2009
[21]

Concordance between logical memory and craft story 21 in community-dwelling older adults: the role of demographic factors and cognitive status,

C. O. Nester, J. Qin, C. Wang, M. J. Katz, R. B. Lipton, and L. A. Rabin, “Concordance between logical memory and craft story 21 in community-dwelling older adults: the role of demographic factors and cognitive status,”Archives of Clinical Neuropsychology, vol. 38, no. 7, pp. 1091–1105, 2023

work page 2023
[22]

Stasenko, D

A. Stasenko, D. M. Jacobs, D. P. Salmon, and T. H. Gollan, “The multilingual naming test (mint) as a measure of picture naming ability SALAZAR CAVAZOS ET . AL.: BEHAVIOR SCORE PREDICTION IN RESTING-ST A TE FUNCTIONAL MRI BY DEEP ST A TE SP ACE MODELING 13 in alzheimer’s disease,”Journal of the International Neuropsychological Society, vol. 25, no. 8, pp. ...

work page 2019
[23]

The brain imaging data structure, a format for organizing and describing outputs of neuroimaging experiments,

K. J. Gorgolewski, T. Auer, V . D. Calhoun, R. C. Craddock, S. Das, E. P. Duff, G. Flandin, S. S. Ghosh, T. Glatard, Y . O. Halchenkoet al., “The brain imaging data structure, a format for organizing and describing outputs of neuroimaging experiments,”Scientific data, vol. 3, no. 1, pp. 1–9, 2016

work page 2016
[24]

fmriprep: a robust preprocessing pipeline for functional mri,

O. Esteban, C. J. Markiewicz, R. W. Blair, C. A. Moodie, A. I. Isik, A. Erramuzpe, J. D. Kent, M. Goncalves, E. DuPre, M. Snyderet al., “fmriprep: a robust preprocessing pipeline for functional mri,”Nature methods, vol. 16, no. 1, pp. 111–116, 2019

work page 2019
[25]

Templateflow: a community archive of imaging tem- plates and atlases for improved consistency in neuroimaging,

R. Ciric, R. Lorenz, W. Thompson, M. Goncalves, E. MacNicol, C. J. Markiewicz, Y . O. Halchenko, S. S. Ghosh, K. J. Gorgolewski, R. A. Poldracket al., “Templateflow: a community archive of imaging tem- plates and atlases for improved consistency in neuroimaging,”bioRxiv, pp. 2021–02, 2021

work page 2021
[26]

Functional network organization of the human brain,

J. D. Power, A. L. Cohen, S. M. Nelson, G. S. Wig, K. A. Barnes, J. A. Church, A. C. V ogel, T. O. Laumann, F. M. Miezin, B. L. Schlaggar et al., “Functional network organization of the human brain,”Neuron, vol. 72, no. 4, pp. 665–678, 2011

work page 2011
[27]

Ridge regression,

G. C. McDonald, “Ridge regression,”Wiley Interdisciplinary Reviews: Computational Statistics, vol. 1, no. 1, pp. 93–100, 2009

work page 2009
[28]

Kernel ridge regression,

V . V ovk, “Kernel ridge regression,” inEmpirical inference: Festschrift in honor of vladimir n. vapnik. Springer, 2013, pp. 105–116

work page 2013
[29]

Sch ¨olkopf and A

B. Sch ¨olkopf and A. J. Smola,Learning with kernels: support vector machines, regularization, optimization, and beyond. MIT press, 2002

work page 2002
[30]

A neuromarker of sustained attention from whole-brain functional connectivity,

M. D. Rosenberg, E. S. Finn, D. Scheinost, X. Papademetris, X. Shen, R. T. Constable, and M. M. Chun, “A neuromarker of sustained attention from whole-brain functional connectivity,”Nature neuroscience, vol. 19, no. 1, pp. 165–171, 2016

work page 2016
[31]

Functional connectome fingerprinting: identifying individuals using patterns of brain connectiv- ity,

E. S. Finn, X. Shen, D. Scheinost, M. D. Rosenberg, J. Huang, M. M. Chun, X. Papademetris, and R. T. Constable, “Functional connectome fingerprinting: identifying individuals using patterns of brain connectiv- ity,”Nature neuroscience, vol. 18, no. 11, pp. 1664–1671, 2015

work page 2015
[32]

Task-induced brain state manipulation improves prediction of individual traits,

A. S. Greene, S. Gao, D. Scheinost, and R. T. Constable, “Task-induced brain state manipulation improves prediction of individual traits,”Nature communications, vol. 9, no. 1, p. 2807, 2018

work page 2018
[33]

Independent component analysis: algorithms and applications,

A. Hyv ¨arinen and E. Oja, “Independent component analysis: algorithms and applications,”Neural networks, vol. 13, no. 4-5, pp. 411–430, 2000

work page 2000
[34]

A method for making group inferences from functional mri data using independent component analysis,

V . D. Calhoun, T. Adali, G. D. Pearlson, and J. J. Pekar, “A method for making group inferences from functional mri data using independent component analysis,”Human brain mapping, vol. 14, no. 3, pp. 140– 151, 2001

work page 2001
[35]

Inves- tigations into resting-state connectivity using independent component analysis,

C. F. Beckmann, M. DeLuca, J. T. Devlin, and S. M. Smith, “Inves- tigations into resting-state connectivity using independent component analysis,”Philosophical Transactions of the Royal Society B: Biological Sciences, vol. 360, no. 1457, pp. 1001–1013, 2005

work page 2005
[36]

Hand classification of fmri ica noise components,

L. Griffanti, G. Douaud, J. Bijsterbosch, S. Evangelisti, F. Alfaro- Almagro, M. F. Glasser, E. P. Duff, S. Fitzgibbon, R. Westphal, D. Caroneet al., “Hand classification of fmri ica noise components,” Neuroimage, vol. 154, pp. 188–205, 2017

work page 2017
[37]

A review of group ica for fmri data and ica for joint inference of imaging, genetic, and erp data,

V . D. Calhoun, J. Liu, and T. Adalı, “A review of group ica for fmri data and ica for joint inference of imaging, genetic, and erp data,” Neuroimage, vol. 45, no. 1, pp. S163–S172, 2009

work page 2009
[38]

Comparison of three methods for generating group statistical inferences from independent component analysis of functional magnetic resonance imaging data,

V . J. Schmithorst and S. K. Holland, “Comparison of three methods for generating group statistical inferences from independent component analysis of functional magnetic resonance imaging data,”Journal of Magnetic Resonance Imaging: An Official Journal of the International Society for Magnetic Resonance in Medicine, vol. 19, no. 3, pp. 365– 368, 2004

work page 2004
[39]

Gradual disturbances of the amplitude of low-frequency fluctuations (alff) and fractional alff in alzheimer spectrum,

L. Yang, Y . Yan, Y . Wang, X. Hu, J. Lu, P. Chan, T. Yan, and Y . Han, “Gradual disturbances of the amplitude of low-frequency fluctuations (alff) and fractional alff in alzheimer spectrum,”Frontiers in neuro- science, vol. 12, p. 975, 2018

work page 2018
[40]

Amplitude of low-frequency fluctuation (alff) and fractional alff in migraine patients: a resting-state functional mri study,

J.-J. Wang, X. Chen, S. Sah, C. Zeng, Y .-M. Li, N. Li, M.-Q. Liu, and S.-L. Du, “Amplitude of low-frequency fluctuation (alff) and fractional alff in migraine patients: a resting-state functional mri study,”Clinical radiology, vol. 71, no. 6, pp. 558–564, 2016

work page 2016
[41]

Frequency-specific alter- nations in the amplitude of low-frequency fluctuations in schizophrenia,

R. Yu, Y .-L. Chien, H.-L. S. Wang, C.-M. Liu, C.-C. Liu, T.-J. Hwang, M. H. Hsieh, H.-G. Hwu, and W.-Y . I. Tseng, “Frequency-specific alter- nations in the amplitude of low-frequency fluctuations in schizophrenia,” Human brain mapping, vol. 35, no. 2, pp. 627–637, 2014

work page 2014
[42]

An empirical evaluation of generic convolutional and recurrent networks for sequence modeling,

S. Bai, J. Z. Kolter, and V . Koltun, “An empirical evaluation of generic convolutional and recurrent networks for sequence modeling,” inThe Sixth International Conference on Learning Representations (ICLR), 2018

work page 2018
[43]

Fundamentals of recurrent neural network (rnn) and long short-term memory (lstm) network,

A. Sherstinsky, “Fundamentals of recurrent neural network (rnn) and long short-term memory (lstm) network,”Physica D: Nonlinear Phe- nomena, vol. 404, p. 132306, 2020

work page 2020
[44]

Bidirectional lstm networks for improved phoneme classification and recognition,

A. Graves, S. Fern ´andez, and J. Schmidhuber, “Bidirectional lstm networks for improved phoneme classification and recognition,” in International conference on artificial neural networks. Springer, 2005, pp. 799–804

work page 2005
[45]

A time series is worth 64 words: Long-term forecasting with transformers,

Y . Nie, N. H. Nguyen, P. Sinthong, and J. Kalagnanam, “A time series is worth 64 words: Long-term forecasting with transformers,” inThe Eleventh International Conference on Learning Representations (ICLR), 2023

work page 2023
[46]

Attention is all you need,

A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin, “Attention is all you need,”Advances in neural information processing systems, vol. 30, 2017

work page 2017
[47]

Aoki,State space modeling of time series

M. Aoki,State space modeling of time series. Springer Science & Business Media, 2013

work page 2013
[48]

Efficiently modeling long sequences with structured state spaces,

A. Gu, K. Goel, and C. Re, “Efficiently modeling long sequences with structured state spaces,” inInternational Conference on Learning Representations, 2022

work page 2022
[49]

Mamba: Linear-time sequence modeling with selective state spaces,

A. Gu and T. Dao, “Mamba: Linear-time sequence modeling with selective state spaces,” inInternational Conference on Learning Rep- resentations, 2024

work page 2024
[50]

Transformers are ssms: Generalized models and effi- cient algorithms through structured state space duality,

T. Dao and A. Gu, “Transformers are ssms: Generalized models and effi- cient algorithms through structured state space duality,” inInternational Conference on Machine Learning. PMLR, 2024, pp. 10 041–10 071

work page 2024
[51]

Prefix sums and their applications,

G. E. Blelloch, “Prefix sums and their applications,”Carnegie Mellon University Pittsburgh, PA, USA, 1990

work page 1990
[52]

Hippo: Recurrent memory with optimal polynomial projections,

A. Gu, T. Dao, S. Ermon, A. Rudra, and C. R ´e, “Hippo: Recurrent memory with optimal polynomial projections,”Advances in neural information processing systems, vol. 33, pp. 1474–1487, 2020

work page 2020
[53]

Differential transformer,

T. Ye, L. Dong, Y . Xia, Y . Sun, Y . Zhu, G. Huang, and F. Wei, “Differential transformer,” inThe Thirteenth International Conference on Learning Representations, 2025

work page 2025
[54]

Differential mamba,

N. Schneider, I. Zimerman, and E. Nachmani, “Differential mamba,” in Association for Computational Linguistics (AACL), 2025

work page 2025
[55]

Implicit Regularization of Stochastic Gradient Descent in Natural Language Processing: Observations and Implications

D. Lei, Z. Sun, Y . Xiao, and W. Y . Wang, “Implicit regularization of stochastic gradient descent in natural language processing: Observations and implications,”arXiv preprint arXiv:1811.00659, 2018

work page internal anchor Pith review Pith/arXiv arXiv 2018
[56]

Revisiting Small Batch Training for Deep Neural Networks

D. Masters and C. Luschi, “Revisiting small batch training for deep neural networks,”arXiv preprint arXiv:1804.07612, 2018

work page internal anchor Pith review Pith/arXiv arXiv 2018
[57]

Sgd: The role of implicit regularization, batch-size and multiple-epochs,

A. Sekhari, K. Sridharan, and S. Kale, “Sgd: The role of implicit regularization, batch-size and multiple-epochs,”Advances In Neural Information Processing Systems, vol. 34, pp. 27 422–27 433, 2021

work page 2021
[58]

Root mean square layer normalization,

B. Zhang and R. Sennrich, “Root mean square layer normalization,” Advances in neural information processing systems, vol. 32, 2019

work page 2019
[59]

Permutation importance: a corrected feature importance measure,

A. Altmann, L. Tolosi, O. Sander, and T. Lengauer, “Permutation importance: a corrected feature importance measure,”Bioinformatics, vol. 26, no. 10, pp. 1340–1347, 2010

work page 2010
[60]

Automated talairach atlas labels for functional brain mapping,

J. L. Lancaster, M. G. Woldorff, L. M. Parsons, M. Liotti, C. S. Freitas, L. Rainey, P. V . Kochunov, D. Nickerson, S. A. Mikiten, and P. T. Fox, “Automated talairach atlas labels for functional brain mapping,”Human brain mapping, vol. 10, no. 3, pp. 120–131, 2000

work page 2000
[61]

The parahippocampal gyrus links the default-mode cortical network with the medial temporal lobe memory system,

A. M. Ward, A. P. Schultz, W. Huijbers, K. R. Van Dijk, T. Hedden, and R. A. Sperling, “The parahippocampal gyrus links the default-mode cortical network with the medial temporal lobe memory system,”Human brain mapping, vol. 35, no. 3, pp. 1061–1073, 2014

work page 2014
[62]

Magnetic resonance imaging of the entorhinal cortex and hippocampus in mild cognitive impairment and alzheimer’s disease,

A. Du, N. Schuff, D. Amend, M. Laakso, Y . Hsu, W. Jagust, K. Yaffe, J. Kramer, B. Reed, D. Normanet al., “Magnetic resonance imaging of the entorhinal cortex and hippocampus in mild cognitive impairment and alzheimer’s disease,”Journal of Neurology, Neurosurgery & Psychiatry, vol. 71, no. 4, pp. 441–447, 2001

work page 2001
[63]

Alterations in memory networks in mild cognitive impairment and alzheimer’s disease: an independent component analysis,

K. A. Celone, V . D. Calhoun, B. C. Dickerson, A. Atri, E. F. Chua, S. L. Miller, K. DePeau, D. M. Rentz, D. J. Selkoe, D. Blackeret al., “Alterations in memory networks in mild cognitive impairment and alzheimer’s disease: an independent component analysis,”Journal of Neuroscience, vol. 26, no. 40, pp. 10 222–10 231, 2006

work page 2006
[64]

Neuropathological stageing of alzheimer-related changes,

H. Braak and E. Braak, “Neuropathological stageing of alzheimer-related changes,”Acta neuropathologica, vol. 82, no. 4, pp. 239–259, 1991

work page 1991
[65]

Default-mode network activity distinguishes alzheimer’s disease from healthy aging: evidence from functional mri,

M. D. Greicius, G. Srivastava, A. L. Reiss, and V . Menon, “Default-mode network activity distinguishes alzheimer’s disease from healthy aging: evidence from functional mri,”Proceedings of the National Academy of Sciences, vol. 101, no. 13, pp. 4637–4642, 2004

work page 2004
[66]

Resting-state fmri changes in alzheimer’s disease and mild cognitive impairment,

M. A. Binnewijzend, M. M. Schoonheim, E. Sanz-Arigita, A. M. Wink, W. M. van der Flier, N. Tolboom, S. M. Adriaanse, J. S. Damoiseaux, P. Scheltens, B. N. van Berckelet al., “Resting-state fmri changes in alzheimer’s disease and mild cognitive impairment,”Neurobiology of aging, vol. 33, no. 9, pp. 2018–2028, 2012. 14

work page 2018
[67]

Study of brain morphology change in alzheimer’s disease and amnestic mild cognitive impairment compared with normal controls,

H. Yang, H. Xu, Q. Li, Y . Jin, W. Jiang, J. Wang, Y . Wu, W. Li, C. Yang, X. Liet al., “Study of brain morphology change in alzheimer’s disease and amnestic mild cognitive impairment compared with normal controls,”General psychiatry, vol. 32, no. 2, p. e100005, 2019

work page 2019
[68]

Cognitive efficiency in alzheimer’s disease is associated with increased occipital connectivity,

M. De Marco, D. Duzzi, F. Meneghello, and A. Venneri, “Cognitive efficiency in alzheimer’s disease is associated with increased occipital connectivity,”Journal of Alzheimer’s Disease, vol. 57, no. 2, pp. 541– 556, 2017

work page 2017
[69]

The precuneus: a review of its functional anatomy and behavioural correlates,

A. E. Cavanna and M. R. Trimble, “The precuneus: a review of its functional anatomy and behavioural correlates,”Brain, vol. 129, no. 3, pp. 564–583, 2006

work page 2006
[70]

Bailly, C

M. Bailly, C. Destrieux, C. Hommet, K. Mondon, J.-P. Cottier, E. Beau- fils, E. Vierron, J. Vercouillie, M. Ibazizene, T. V oisinet al., “Precuneus and cingulate cortex atrophy and hypometabolism in patients with alzheimer’s disease and mild cognitive impairment: Mri and 18f-fdg pet quantitative analysis using freesurfer,”BioMed research international, vo...

work page 2015
[71]

Precuneus atrophy in early-onset alzheimer’s disease: a morphometric structural mri study,

G. Karas, P. Scheltens, S. Rombouts, R. Van Schijndel, M. Klein, B. Jones, W. Van Der Flier, H. Vrenken, and F. Barkhof, “Precuneus atrophy in early-onset alzheimer’s disease: a morphometric structural mri study,”Neuroradiology, vol. 49, no. 12, pp. 967–976, 2007

work page 2007
[72]

On the neuroanatomy and functional role of the inferior parietal lobule and intraparietal sulcus,

F. C. Binkofski, J. Klann, and S. Caspers, “On the neuroanatomy and functional role of the inferior parietal lobule and intraparietal sulcus,” inNeurobiology of language. Elsevier, 2016, pp. 35–47

work page 2016
[73]

Subregions of the inferior parietal lobule are affected in the progression to alzheimer’s disease,

S. J. Greene, R. J. Killiany, A. D. N. Initiativeet al., “Subregions of the inferior parietal lobule are affected in the progression to alzheimer’s disease,”Neurobiology of aging, vol. 31, no. 8, pp. 1304–1311, 2010

work page 2010
[74]

Differentially disrupted functional connectivity of the subregions of the inferior parietal lobule in alzheimer’s disease,

Z. Wang, M. Xia, Z. Dai, X. Liang, H. Song, Y . He, and K. Li, “Differentially disrupted functional connectivity of the subregions of the inferior parietal lobule in alzheimer’s disease,”Brain Structure and Function, vol. 220, no. 2, pp. 745–762, 2015

work page 2015
[75]

Anterior cingulate cortex: unique role in cognition and emotion,

F. L. Stevens, R. A. Hurley, and K. H. Taber, “Anterior cingulate cortex: unique role in cognition and emotion,”The Journal of neuropsychiatry and clinical neurosciences, vol. 23, no. 2, pp. 121–125, 2011

work page 2011
[76]

Reduced thickness of the anterior cingulate cortex as a predictor of amnestic-mild cognitive impairment conversion to alzheimer’s disease with psychosis,

H.-J. Jeong, Y .-M. Lee, J.-M. Park, B.-D. Lee, E. Moon, H. Suh, H.- J. Kim, K. Pak, K.-U. Choi, and Y .-I. Chung, “Reduced thickness of the anterior cingulate cortex as a predictor of amnestic-mild cognitive impairment conversion to alzheimer’s disease with psychosis,”Journal of Alzheimer’s Disease, vol. 84, no. 4, pp. 1709–1717, 2021

work page 2021
[77]

Decreased functional connectivity between the dorsal anterior cingulate cortex and lingual gyrus in alzheimer’s disease patients with depression,

X. Liu, W. Chen, H. Hou, X. Chen, J. Zhang, J. Liu, Z. Guo, and G. Bai, “Decreased functional connectivity between the dorsal anterior cingulate cortex and lingual gyrus in alzheimer’s disease patients with depression,”Behavioural brain research, vol. 326, pp. 132–138, 2017

work page 2017
[78]

Orbitofrontal and anterior cingulate cortex neurofibrillary tangle burden is associated with agitation in alzheimer disease,

S. Tekin, M. S. Mega, D. M. Masterman, T. Chow, J. Garakian, H. V . Vinters, and J. L. Cummings, “Orbitofrontal and anterior cingulate cortex neurofibrillary tangle burden is associated with agitation in alzheimer disease,”Annals of neurology, vol. 49, no. 3, pp. 355–361, 2001

work page 2001
[79]

Alzheimers disease classification in functional mri with 4d joint temporal-spatial kernels in novel 4d cnn model,

J. Salazar Cavazos and S. Peltier, “Alzheimers disease classification in functional mri with 4d joint temporal-spatial kernels in novel 4d cnn model,”arXiv preprint arXiv:2506.02060, 2025

work page arXiv 2025
[80]

A comprehensive analysis of resting state fmri measures to classify individual patients with alzheimer’s disease,

F. de V os, M. Koini, T. M. Schouten, S. Seiler, J. van der Grond, A. Lech- ner, R. Schmidt, M. de Rooij, and S. A. Rombouts, “A comprehensive analysis of resting state fmri measures to classify individual patients with alzheimer’s disease,”Neuroimage, vol. 167, pp. 62–72, 2018

work page 2018
[81]

Multi-label classification of alzheimer’s disease stages from resting-state fmri-based correlation connectivity data and deep learning,

A. Alorf and M. U. G. Khan, “Multi-label classification of alzheimer’s disease stages from resting-state fmri-based correlation connectivity data and deep learning,”Computers in Biology and Medicine, vol. 151, p. 106240, 2022

work page 2022

Showing first 80 references.