Medication-Aware Financial Exploitation Detection for Alzheimer's Patients Using Edge-Aware Interaction Risk Modeling

Chaitanya Gunupudi; Faisal Quader; Farzana Akter; Lisan Al Amin; Rakib Hossain

arxiv: 2606.00672 · v1 · pith:QQRJ6DEFnew · submitted 2026-05-30 · 💻 cs.AI · cs.LG

Medication-Aware Financial Exploitation Detection for Alzheimer's Patients Using Edge-Aware Interaction Risk Modeling

Farzana Akter , Lisan Al Amin , Rakib Hossain , Chaitanya Gunupudi , Faisal Quader This is my paper

Pith reviewed 2026-06-28 18:38 UTC · model grok-4.3

classification 💻 cs.AI cs.LG

keywords financial exploitationAlzheimer's diseasemedication adherencefraud detectioninteraction modelingcognitive vulnerabilitysimulation datasetlogistic regression

0 comments

The pith

Medication adherence data, when used as a contextual modifier in an interaction-aware model, improves recall of financial exploitation risks in Alzheimer's patients from 0.7442 to 0.9070 during vulnerable windows.

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

The paper builds a medication-aware detection framework that links medication adherence records to individual financial transactions for people with Alzheimer's disease. It creates a simulated dataset covering 180 patients over 45 days and tests three logistic models: one based only on transaction features such as amount anomaly and frequency, one that adds medication data directly, and one that models explicit interactions between medication and financial signals. The interaction version raises recall specifically inside medication-induced vulnerability periods and ranks high-risk cases with the highest average precision, even though the financial-only version scores highest on overall F1. A sympathetic reader would care because standard fraud systems ignore how medication timing can temporarily raise exploitation risk, and the results point to medication adherence working best as a risk modifier rather than a separate signal.

Core claim

The interaction-aware logistic model that treats medication adherence as a modifier of financial features such as amount anomaly, vendor novelty, transaction frequency, and time deviation improves recall during medication-induced vulnerability windows from 0.7442 to 0.9070 while achieving the highest average precision for ranked high-risk cases; the financial-only baseline reaches the highest global F1-score of 0.5000. The findings indicate that medication adherence functions most effectively as a contextual modifier of financial risk rather than as an isolated predictor.

What carries the argument

The interaction-aware logistic model that combines transaction features with medication adherence to capture how medication alters financial risk patterns.

If this is right

The interaction-aware model delivers higher recall than baselines inside medication-linked vulnerability windows.
Medication adherence data improves ranking of high-risk financial cases when treated as a risk modifier.
Financial-only models miss periods of elevated exploitation risk tied to medication effects.
The framework can evaluate model variants on simulated data without requiring real combined records.

Where Pith is reading between the lines

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

Banking systems could incorporate pharmacy adherence feeds to trigger alerts only during predicted high-vulnerability intervals.
The same modifier approach might extend to other temporary cognitive states such as post-surgical medication effects.
Deployment would require privacy-preserving linkage between clinical and financial data sources.
Adding dosage timing or specific drug classes could sharpen the vulnerability windows further.

Load-bearing premise

The hybrid simulation dataset accurately reflects the real relationships between medication adherence and financial vulnerability windows in Alzheimer's patients.

What would settle it

A validation study on actual patient transaction and medication records that finds no recall or precision gain from the interaction terms would show the claimed advantage does not hold.

Figures

Figures reproduced from arXiv: 2606.00672 by Chaitanya Gunupudi, Faisal Quader, Farzana Akter, Lisan Al Amin, Rakib Hossain.

**Figure 1.** Figure 1: Proposed medication-aware financial exploitation detection framework. Medication adherence and transaction streams are temporally synchronized to construct clinically informed financial risk representations. The framework compares baseline, additive medication-aware, and interaction-aware models, then produces vulnerabilitysensitive risk scores and prioritized safeguard actions in a simulation-based, edg… view at source ↗

**Figure 2.** Figure 2: Precision-recall curves for financial-only, additive medication-aware, and interaction-aware models. The interaction-aware model achieves the highest average precision, indicating improved ranking quality for high-risk exploitation cases. recall gain is important because missed exploitation events during cognitively vulnerable periods are more harmful than additional cases sent for review. The result shows… view at source ↗

read the original abstract

Financial exploitation is a growing concern for people with Alzheimer's disease, especially during periods of reduced cognitive stability. Conventional fraud detection systems usually rely on financial behavior alone and ignore clinically relevant factors that may alter vulnerability. This paper proposes a medication-aware framework that synchronizes medication adherence with transaction-level monitoring to improve detection of cognitively risky financial events. A hybrid simulation dataset was constructed for 180 patients across 45 days, producing 8,100 medication records and 30,855 transactions. The framework evaluates amount anomaly, vendor novelty, transaction frequency, time deviation, and medication adherence through financial-only, additive medication-aware, and interaction-aware logistic models. Results show that the financial-only baseline obtained the highest global F1-score of 0.5000, but the interaction-aware model improved recall during medication-induced vulnerability windows from 0.7442 to 0.9070 and achieved the highest average precision for ranked high-risk cases. The findings suggest that medication adherence is most useful as a contextual modifier of financial risk rather than as an isolated predictor.

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 reported recall lift from 0.74 to 0.91 occurs only inside an unspecified hybrid simulation whose rules for linking medication to transaction labels are not described.

read the letter

The paper's main result is that an interaction-aware logistic regression improves recall during medication-induced vulnerability windows on their simulated data, while the plain financial model scores highest on overall F1. They built a hybrid dataset for 180 patients over 45 days and compared three logistic models on features like amount anomaly, vendor novelty, and medication adherence.

The domain combination is straightforward: financial transaction patterns plus medication adherence as a contextual factor for Alzheimer's patients. The three-model comparison is clear enough on the surface, and the interaction term does show the targeted lift plus better average precision for high-risk ranking.

The soft spot is the simulation itself. The abstract says the dataset was constructed but supplies no equations, parameters, or validation steps for how adherence was turned into vulnerability labels or altered transaction patterns. All metrics come from models fit to the same data that generated the labels, with no hold-out, cross-validation details, or external cohort mentioned. This leaves the numeric improvement tied to whatever generative assumptions were used.

The work targets applied researchers in fraud detection or digital health who might consider adding clinical context to transaction monitoring. It does not supply enough methodological transparency or external grounding to support the claims or allow others to reproduce or extend the result.

I would not bring this to a reading group or cite it. It does not deserve peer review because the central performance claims rest on an unvalidated simulation without the details needed to assess whether the improvement is real or an artifact of the data construction.

Referee Report

3 major / 1 minor

Summary. The paper proposes a medication-aware framework for detecting financial exploitation in Alzheimer's patients by synchronizing medication adherence data with transaction monitoring. It constructs a hybrid simulation dataset for 180 patients over 45 days (8,100 medication records, 30,855 transactions) and compares three logistic regression models—financial-only, additive medication-aware, and interaction-aware—on features including amount anomaly, vendor novelty, transaction frequency, time deviation, and medication adherence. The central claim is that the interaction-aware model improves recall during medication-induced vulnerability windows from 0.7442 to 0.9070 and achieves the highest average precision for ranked high-risk cases, while the financial-only baseline has the highest global F1 of 0.5000; medication adherence is positioned as a contextual modifier rather than an isolated predictor.

Significance. If the simulation's generative rules for medication-vulnerability interactions can be shown to be realistic and non-circular, the result would indicate that clinical factors can usefully modulate financial risk models for cognitively vulnerable populations, supporting more targeted fraud detection in healthcare-adjacent financial systems.

major comments (3)

[Abstract] Abstract and results paragraph: the hybrid simulation is described only as 'constructed' for 180 patients; no equations, parameters, generative rules, or validation steps are supplied for mapping medication adherence onto transaction patterns or ground-truth vulnerability labels. Without these, the reported recall lift (0.7442 → 0.9070) cannot be distinguished from an artifact of the data-generation process that directly encodes the interaction signal the model is designed to exploit.
[Results paragraph] Results paragraph: all metrics (recall, average precision, F1) are obtained from logistic models whose coefficients were fitted to the identical simulated dataset that produced the vulnerability windows and labels; no train-test split, cross-validation procedure, or external hold-out cohort is mentioned, so the numeric improvements are tied directly to the fitted quantities rather than to generalization.
[Abstract] Abstract: the financial-only baseline reports the highest global F1-score (0.5000), yet emphasis is placed on subset performance within medication-induced windows; this selective focus requires explicit justification of why overall utility is not the primary evaluation criterion when claiming practical value for exploitation detection.

minor comments (1)

The abstract lists features (amount anomaly, vendor novelty, etc.) but does not define their exact computation or how they enter the three logistic models; adding a methods subsection with feature definitions would improve reproducibility.

Simulated Author's Rebuttal

3 responses · 0 unresolved

Thank you for the referee's constructive comments, which identify key areas where the manuscript requires greater transparency on the simulation process, evaluation methodology, and metric prioritization. We address each point below and commit to revisions that strengthen the paper without altering its core claims.

read point-by-point responses

Referee: [Abstract] Abstract and results paragraph: the hybrid simulation is described only as 'constructed' for 180 patients; no equations, parameters, generative rules, or validation steps are supplied for mapping medication adherence onto transaction patterns or ground-truth vulnerability labels. Without these, the reported recall lift (0.7442 → 0.9070) cannot be distinguished from an artifact of the data-generation process that directly encodes the interaction signal the model is designed to exploit.

Authors: We agree that the current description of the simulation is insufficient. The revised manuscript will include the full generative equations, all parameter values, the precise rules linking medication adherence to vulnerability windows and transaction anomalies, and any validation steps used to ensure the labels reflect the intended interactions. This addition will enable readers to assess whether the reported improvements arise from non-circular modeling choices. revision: yes
Referee: [Results paragraph] Results paragraph: all metrics (recall, average precision, F1) are obtained from logistic models whose coefficients were fitted to the identical simulated dataset that produced the vulnerability windows and labels; no train-test split, cross-validation procedure, or external hold-out cohort is mentioned, so the numeric improvements are tied directly to the fitted quantities rather than to generalization.

Authors: The referee is correct that no train-test split or cross-validation procedure is described. The current results reflect in-sample fitting. In the revision we will add an explicit hold-out evaluation (e.g., 70/30 split or k-fold cross-validation) and report performance on unseen data to demonstrate generalization. revision: yes
Referee: [Abstract] Abstract: the financial-only baseline reports the highest global F1-score (0.5000), yet emphasis is placed on subset performance within medication-induced windows; this selective focus requires explicit justification of why overall utility is not the primary evaluation criterion when claiming practical value for exploitation detection.

Authors: We will expand the abstract and add a dedicated paragraph in the discussion to justify the focus on vulnerability-window recall. The justification rests on the clinical priority of minimizing missed exploitation events during periods of elevated risk, where the cost of false negatives is disproportionately high; global F1 will remain reported as a secondary metric for overall system comparison. revision: yes

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the unverified fidelity of the simulation and on the assumption that logistic regression with hand-chosen interaction terms can capture medication-induced vulnerability; no external benchmarks or real patient data are used.

free parameters (1)

logistic regression coefficients
All model weights are fitted to the 30,855 simulated transactions and 8,100 medication records to produce the reported F1, recall, and precision values.

axioms (1)

domain assumption Logistic regression is an appropriate model for binary financial-exploitation risk given the chosen features
The abstract invokes three logistic models without stating why this family was selected over alternatives or how feature interactions were chosen.

pith-pipeline@v0.9.1-grok · 5727 in / 1391 out tokens · 29307 ms · 2026-06-28T18:38:35.250127+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

21 extracted references · 16 canonical work pages

[1]

In: SoutheastCon 2026

Akter, F., Hossain, R., Toushi, D.K.R., Khan, M.M., Amin, S., Al Amin, L.: Hybrid federated and split learning for privacy preserving clinical prediction and treatment optimization. In: SoutheastCon 2026. pp. 1–8. IEEE (2026)

2026
[2]

IEEE Access11, 396–409 (2022)

Al Amin, L., Mukta, M.S.H., Saikat, M.S.M., Hossain, M.I., Islam, M.A., Ahmed, M., Azam, S.: Data driven classification of opioid patients using machine learning– an investigation. IEEE Access11, 396–409 (2022)

2022
[3]

In: 2025 IEEE International Conference on Data Mining Workshops (ICDMW)

Al Mamun, S., Hossain, R., Rahman, M.J., Devnath, M.K., Afroz, F., Al Amin, L.: Bayesian modeling for uncertainty management in financial risk forecasting and compliance. In: 2025 IEEE International Conference on Data Mining Workshops (ICDMW). pp. 178–186. IEEE (2025)

2025
[4]

https://doi.org/10.4108/eetpht.11.6170

Anik, F.I., Hasan, M.M., Rodriguez-Cardenas, J., Beiswanger, R., Tasnim, M., Ramos,M.,Sakib,N.:IoMTanddataprivacyinalzheimer’scareforolderadults:A systematicreview.EAIEndorsedTransactionsonPervasiveHealthandTechnology 11(1) (2025). https://doi.org/10.4108/eetpht.11.6170

work page doi:10.4108/eetpht.11.6170 2025
[5]

Bazgir, A., Habibdoust, A., Song, X., Zhang, Y.: Agenticad: A specialized mul- tiagent system framework for holistic alzheimer disease management (2025), https://arxiv.org/abs/2510.08578

work page arXiv 2025
[6]

Elec- tronics14(17), 3516 (2025)

Dall’Ora, N., Felli, L., Aldegheri, S., Vicino, N., Giuliano, R.: Lumicare: A context- aware mobile system for alzheimer’s patients integrating AI agents and 6G. Elec- tronics14(17), 3516 (2025). https://doi.org/10.3390/electronics14173516

work page doi:10.3390/electronics14173516 2025
[7]

Advances in Geriatric Medicine and Research5(3), e230007 (2023)

Ebner, N.C., Pehlivanoglu, D., Shoenfelt, A.: Financial fraud and deception in aging. Advances in Geriatric Medicine and Research5(3), e230007 (2023). https://doi.org/10.20900/agmr20230007

work page doi:10.20900/agmr20230007 2023
[8]

Brain Sci- ences14(11), 1113 (2024)

Giannouli, V.: Can changes in financial performance be used in the diagnosis of neurocognitive disorders? a systematic review of findings from greece. Brain Sci- ences14(11), 1113 (2024). https://doi.org/10.3390/brainsci14111113

work page doi:10.3390/brainsci14111113 2024
[9]

Frontiers in Aging Neuroscience17, 1735892 (2026)

Grammenos, G., Vrahatis, A.G., Lazaros, K., Exarchos, T.P., Vlamos, P., Krokidis, M.G.: Ai agents in alzheimer’s disease management: Challenges and future directions. Frontiers in Aging Neuroscience17, 1735892 (2026). https://doi.org/10.3389/fnagi.2025.1735892

work page doi:10.3389/fnagi.2025.1735892 2026
[10]

In: 2025 40th International Conference on Image and Vision Computing New Zealand (IVCNZ)

Hossain,R.,Ali,L.E.,Ripon,K.S.N.:Risk-controlledmultimodalemotioncoaching for autism support using self-supervised vision and speech encoders. In: 2025 40th International Conference on Image and Vision Computing New Zealand (IVCNZ). pp. 1–7. IEEE (2025)

2025
[11]

In: 2025 28th International Conference on Computer and Information Technology (ICCIT)

Hossain, R., Ali, L.E., Ripon, K.S.N., Bulbul, M.F., Tonmoy, M.M.H.: Profit- optimal machine learning for purchase intent prediction: Calibration, temporal validation, and economic constraints. In: 2025 28th International Conference on Computer and Information Technology (ICCIT). pp. 3087–3092. IEEE (2025)

2025
[12]

Journal of Political Economy132(6), 1793–1830 (2024)

Mazzonna, F., Peracchi, F.: Are older people aware of their cognitive decline? misperception and financial decision-making. Journal of Political Economy132(6), 1793–1830 (2024). https://doi.org/10.1086/728697

work page doi:10.1086/728697 2024
[13]

BioMedical Engineering OnLine22(1), 4 (2023)

Mishra, P.K., Iaboni, A., Ye, B., Newman, K., Mihailidis, A., Khan, S.S.: Privacy- protecting behaviours of risk detection in people with dementia using videos. BioMedical Engineering OnLine22(1), 4 (2023). https://doi.org/10.1186/s12938- 023-01065-3 14 F. Akter et al

work page doi:10.1186/s12938- 2023
[14]

Alzheimer’s & Dementia19(4), 1184–1193 (2023)

Olchanski, N., Daly, A.T., Zhu, Y., Breslau, R., Cohen, J.T., Neumann, P.J., et al.: Alzheimer’s disease medication use and adherence patterns by race and ethnicity. Alzheimer’s & Dementia19(4), 1184–1193 (2023). https://doi.org/10.1002/alz.12753

work page doi:10.1002/alz.12753 2023
[15]

JMIR Aging8, e69175 (2025)

Shaik, M.A., Anik, F.I., Hasan, M.M., Chakravarty, S., Ramos, M.D., Rahman, M.A., Ahamed, S.I., Sakib, N.: Advancing remote monitoring for patients with alzheimer disease and related dementias: Systematic review. JMIR Aging8, e69175 (2025). https://doi.org/10.2196/69175

work page doi:10.2196/69175 2025
[16]

JAMA Network Open8(6), e2515894 (2025)

Trendl, A., Anwyl-Irvine, A., Vomfell, L., Abbey, E., Stewart, N., Atkins, D., Llewellyn, D.J., Gathergood, J., Leake, D.: Early behavioral markers of loss of financial capacity. JAMA Network Open8(6), e2515894 (2025). https://doi.org/10.1001/jamanetworkopen.2025.15894

work page doi:10.1001/jamanetworkopen.2025.15894 2025
[17]

BioMedInformatics5(3), 49 (2025)

UdDin,F.,Giri,N.,Shetty,N.,Hilton,T.,Shafiabady,N.,Tully,P.J.:Co-designing a DSM-5-based AI-powered smart assistant for monitoring dementia and ongoing neurocognitive decline: Development study. BioMedInformatics5(3), 49 (2025). https://doi.org/10.3390/biomedinformatics5030049

work page doi:10.3390/biomedinformatics5030049 2025
[18]

Trauma, Violence, & Abuse (2025)

Wei, W., Balser, S.: A scoping review: Financial exploitation among older people living with dementia. Trauma, Violence, & Abuse (2025). https://doi.org/10.1177/15248380251383930

work page doi:10.1177/15248380251383930 2025
[19]

Proceedings of the ACM on Inter- active, Mobile, Wearable and Ubiquitous Technologies8(2), 73:1–73:35 (2024)

Yang, Z., Xu, X., Yao, B., Rogers, E., Zhang, S., Intille, S.S., Shara, N., Gao, G.G., Wang, D.: Talk2care: An LLM-based voice assistant for communication be- tween healthcare providers and older adults. Proceedings of the ACM on Inter- active, Mobile, Wearable and Ubiquitous Technologies8(2), 73:1–73:35 (2024). https://doi.org/10.1145/3659625

work page doi:10.1145/3659625 2024
[20]

Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems pp

Zhai, Y., Xue, X., Guo, Z., Jin, T., Diao, Y., Jeung, J.: Hear us, then protect us: Navigating deepfake scams and safeguard interventions with older adults through participatory design. Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems pp. 1–19 (2025). https://doi.org/10.1145/3706598.3714423

work page doi:10.1145/3706598.3714423 2025
[21]

In: Proceedings of the 2024 Conference of the North American Chapter of the Association for Compu- tational Linguistics: Human Language Technologies (Volume 1: Long Papers)

Zhang, K., Kang, Y., Zhao, F., Liu, X.: LLM-based medical assistant personal- ization with short- and long-term memory coordination. In: Proceedings of the 2024 Conference of the North American Chapter of the Association for Compu- tational Linguistics: Human Language Technologies (Volume 1: Long Papers). pp. 2386–2398.AssociationforComputationalLinguisti...

work page doi:10.18653/v1/2024.naacl-long.132 2024

[1] [1]

In: SoutheastCon 2026

Akter, F., Hossain, R., Toushi, D.K.R., Khan, M.M., Amin, S., Al Amin, L.: Hybrid federated and split learning for privacy preserving clinical prediction and treatment optimization. In: SoutheastCon 2026. pp. 1–8. IEEE (2026)

2026

[2] [2]

IEEE Access11, 396–409 (2022)

Al Amin, L., Mukta, M.S.H., Saikat, M.S.M., Hossain, M.I., Islam, M.A., Ahmed, M., Azam, S.: Data driven classification of opioid patients using machine learning– an investigation. IEEE Access11, 396–409 (2022)

2022

[3] [3]

In: 2025 IEEE International Conference on Data Mining Workshops (ICDMW)

Al Mamun, S., Hossain, R., Rahman, M.J., Devnath, M.K., Afroz, F., Al Amin, L.: Bayesian modeling for uncertainty management in financial risk forecasting and compliance. In: 2025 IEEE International Conference on Data Mining Workshops (ICDMW). pp. 178–186. IEEE (2025)

2025

[4] [4]

https://doi.org/10.4108/eetpht.11.6170

Anik, F.I., Hasan, M.M., Rodriguez-Cardenas, J., Beiswanger, R., Tasnim, M., Ramos,M.,Sakib,N.:IoMTanddataprivacyinalzheimer’scareforolderadults:A systematicreview.EAIEndorsedTransactionsonPervasiveHealthandTechnology 11(1) (2025). https://doi.org/10.4108/eetpht.11.6170

work page doi:10.4108/eetpht.11.6170 2025

[5] [5]

Bazgir, A., Habibdoust, A., Song, X., Zhang, Y.: Agenticad: A specialized mul- tiagent system framework for holistic alzheimer disease management (2025), https://arxiv.org/abs/2510.08578

work page arXiv 2025

[6] [6]

Elec- tronics14(17), 3516 (2025)

Dall’Ora, N., Felli, L., Aldegheri, S., Vicino, N., Giuliano, R.: Lumicare: A context- aware mobile system for alzheimer’s patients integrating AI agents and 6G. Elec- tronics14(17), 3516 (2025). https://doi.org/10.3390/electronics14173516

work page doi:10.3390/electronics14173516 2025

[7] [7]

Advances in Geriatric Medicine and Research5(3), e230007 (2023)

Ebner, N.C., Pehlivanoglu, D., Shoenfelt, A.: Financial fraud and deception in aging. Advances in Geriatric Medicine and Research5(3), e230007 (2023). https://doi.org/10.20900/agmr20230007

work page doi:10.20900/agmr20230007 2023

[8] [8]

Brain Sci- ences14(11), 1113 (2024)

Giannouli, V.: Can changes in financial performance be used in the diagnosis of neurocognitive disorders? a systematic review of findings from greece. Brain Sci- ences14(11), 1113 (2024). https://doi.org/10.3390/brainsci14111113

work page doi:10.3390/brainsci14111113 2024

[9] [9]

Frontiers in Aging Neuroscience17, 1735892 (2026)

Grammenos, G., Vrahatis, A.G., Lazaros, K., Exarchos, T.P., Vlamos, P., Krokidis, M.G.: Ai agents in alzheimer’s disease management: Challenges and future directions. Frontiers in Aging Neuroscience17, 1735892 (2026). https://doi.org/10.3389/fnagi.2025.1735892

work page doi:10.3389/fnagi.2025.1735892 2026

[10] [10]

In: 2025 40th International Conference on Image and Vision Computing New Zealand (IVCNZ)

Hossain,R.,Ali,L.E.,Ripon,K.S.N.:Risk-controlledmultimodalemotioncoaching for autism support using self-supervised vision and speech encoders. In: 2025 40th International Conference on Image and Vision Computing New Zealand (IVCNZ). pp. 1–7. IEEE (2025)

2025

[11] [11]

In: 2025 28th International Conference on Computer and Information Technology (ICCIT)

Hossain, R., Ali, L.E., Ripon, K.S.N., Bulbul, M.F., Tonmoy, M.M.H.: Profit- optimal machine learning for purchase intent prediction: Calibration, temporal validation, and economic constraints. In: 2025 28th International Conference on Computer and Information Technology (ICCIT). pp. 3087–3092. IEEE (2025)

2025

[12] [12]

Journal of Political Economy132(6), 1793–1830 (2024)

Mazzonna, F., Peracchi, F.: Are older people aware of their cognitive decline? misperception and financial decision-making. Journal of Political Economy132(6), 1793–1830 (2024). https://doi.org/10.1086/728697

work page doi:10.1086/728697 2024

[13] [13]

BioMedical Engineering OnLine22(1), 4 (2023)

Mishra, P.K., Iaboni, A., Ye, B., Newman, K., Mihailidis, A., Khan, S.S.: Privacy- protecting behaviours of risk detection in people with dementia using videos. BioMedical Engineering OnLine22(1), 4 (2023). https://doi.org/10.1186/s12938- 023-01065-3 14 F. Akter et al

work page doi:10.1186/s12938- 2023

[14] [14]

Alzheimer’s & Dementia19(4), 1184–1193 (2023)

Olchanski, N., Daly, A.T., Zhu, Y., Breslau, R., Cohen, J.T., Neumann, P.J., et al.: Alzheimer’s disease medication use and adherence patterns by race and ethnicity. Alzheimer’s & Dementia19(4), 1184–1193 (2023). https://doi.org/10.1002/alz.12753

work page doi:10.1002/alz.12753 2023

[15] [15]

JMIR Aging8, e69175 (2025)

Shaik, M.A., Anik, F.I., Hasan, M.M., Chakravarty, S., Ramos, M.D., Rahman, M.A., Ahamed, S.I., Sakib, N.: Advancing remote monitoring for patients with alzheimer disease and related dementias: Systematic review. JMIR Aging8, e69175 (2025). https://doi.org/10.2196/69175

work page doi:10.2196/69175 2025

[16] [16]

JAMA Network Open8(6), e2515894 (2025)

Trendl, A., Anwyl-Irvine, A., Vomfell, L., Abbey, E., Stewart, N., Atkins, D., Llewellyn, D.J., Gathergood, J., Leake, D.: Early behavioral markers of loss of financial capacity. JAMA Network Open8(6), e2515894 (2025). https://doi.org/10.1001/jamanetworkopen.2025.15894

work page doi:10.1001/jamanetworkopen.2025.15894 2025

[17] [17]

BioMedInformatics5(3), 49 (2025)

UdDin,F.,Giri,N.,Shetty,N.,Hilton,T.,Shafiabady,N.,Tully,P.J.:Co-designing a DSM-5-based AI-powered smart assistant for monitoring dementia and ongoing neurocognitive decline: Development study. BioMedInformatics5(3), 49 (2025). https://doi.org/10.3390/biomedinformatics5030049

work page doi:10.3390/biomedinformatics5030049 2025

[18] [18]

Trauma, Violence, & Abuse (2025)

Wei, W., Balser, S.: A scoping review: Financial exploitation among older people living with dementia. Trauma, Violence, & Abuse (2025). https://doi.org/10.1177/15248380251383930

work page doi:10.1177/15248380251383930 2025

[19] [19]

Proceedings of the ACM on Inter- active, Mobile, Wearable and Ubiquitous Technologies8(2), 73:1–73:35 (2024)

Yang, Z., Xu, X., Yao, B., Rogers, E., Zhang, S., Intille, S.S., Shara, N., Gao, G.G., Wang, D.: Talk2care: An LLM-based voice assistant for communication be- tween healthcare providers and older adults. Proceedings of the ACM on Inter- active, Mobile, Wearable and Ubiquitous Technologies8(2), 73:1–73:35 (2024). https://doi.org/10.1145/3659625

work page doi:10.1145/3659625 2024

[20] [20]

Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems pp

Zhai, Y., Xue, X., Guo, Z., Jin, T., Diao, Y., Jeung, J.: Hear us, then protect us: Navigating deepfake scams and safeguard interventions with older adults through participatory design. Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems pp. 1–19 (2025). https://doi.org/10.1145/3706598.3714423

work page doi:10.1145/3706598.3714423 2025

[21] [21]

In: Proceedings of the 2024 Conference of the North American Chapter of the Association for Compu- tational Linguistics: Human Language Technologies (Volume 1: Long Papers)

Zhang, K., Kang, Y., Zhao, F., Liu, X.: LLM-based medical assistant personal- ization with short- and long-term memory coordination. In: Proceedings of the 2024 Conference of the North American Chapter of the Association for Compu- tational Linguistics: Human Language Technologies (Volume 1: Long Papers). pp. 2386–2398.AssociationforComputationalLinguisti...

work page doi:10.18653/v1/2024.naacl-long.132 2024