Draft — Africa Healthcare Innovation Fellowship (AHIF)

MOTIVATION LETTER

The CCT model, a tripartite pharmacological framework for preventing reward-memory encoding in addiction, addresses a healthcare gap that African health systems have not yet named as a priority. Substance use disorders across Nigeria and the continent remain under-diagnosed, under-treated, and under-funded, while the pharmacological mechanisms that drive relapse are well-characterized in the global literature. My work builds a computational bridge between those mechanisms and clinical practice. The Africa Healthcare Innovation Fellowship is the first programme I have found that explicitly rewards this kind of context-relevant, field-testable intervention rather than requiring a laboratory or a clinical trial infrastructure I do not have.

The model is a calibrated, tested system. I built a three-axis ODE model coupling dopaminergic reward prediction error, NMDAR-dependent long-term potentiation, and affective contrast, then calibrated all 14 free parameters using Bayesian MCMC with priors elicited from a systematic screen of 1,847 records. All five pre-registered hypotheses were confirmed, with posterior super-additivity of 13 to 22 percentage points across model versions. Three sole-authored preprints are under review at peer-reviewed journals. What the model now needs is a deployment pathway: a screening tool or clinical decision-support prototype that a Nigerian clinician can use to assess relapse risk and guide medication choice. That is precisely the hands-on, field-based work AHIF supports.

My background fits the fellowship's interdisciplinary mandate. I hold a B.Pharm from the University of Ibadan with a German equivalent grade of 1.9, I am a licensed pharmacist, and I am enrolled in the M.Sc. Digital Health programme at the Hasso Plattner Institute in Potsdam. I have built and validated computational pipelines across addiction neuroscience, protein machine learning, and dynamical systems. I have also worked as a clinical pharmacist and as a research assistant on antimicrobial resistance genomics. I know how to read a prescription and how to write a simulation. Few applicants to this fellowship will bring both.

What I need from AHIF is mentorship and a field network to translate a validated computational model into a tool that works in a Nigerian clinic. The 14-week field-based structure is the correct format for this: I need to sit with clinicians, understand their workflow, and adapt the CCT screening instrument to real patient data. The fellowship's emphasis on scalable, context-relevant solutions matches the CCT model's design, which runs on a laptop and requires no specialized hardware.

I am applying to AHIF because it is the only programme I have identified that combines mentorship, field placement, and a focus on African healthcare innovation without requiring me to abandon my computational methods or relocate permanently. The model is ready. The clinical context is ready. The fellowship is the missing link.

RESEARCH STATEMENT

The Conjunctive Consolidation Threshold (CCT) model is a computational framework for understanding why some individuals transition from drug use to addiction and, more importantly, how that transition can be pharmacologically interrupted. The core thesis is that reward-memory consolidation requires the simultaneous activation of three coupled systems: dopaminergic reward prediction error signaling, NMDAR-dependent long-term potentiation in reward circuitry, and affective contrast between drug and non-drug states. Interrupting any one axis is insufficient; the threshold for consolidation is conjunctive.

I operationalized this thesis as a system of ordinary differential equations with three coupled axes, solved using RK45 integration. The model has 14 free parameters, all calibrated using Bayesian MCMC with the DEMetropolisZ sampler in PyMC. Priors were elicited from a systematic literature screen of 1,847 records covering dopaminergic signaling, NMDAR pharmacology, and affective neuroscience. The model was pre-registered with five hypotheses before calibration. All five were confirmed. Posterior analysis showed super-additive effects of multi-axis intervention, ranging from 13 to 22 percentage points over single-axis approaches across model versions. Three sole-authored preprints describing the model, its calibration, and its pharmacological implications are currently under review at peer-reviewed journals. A co-authored paper is under review at Alcohol (Elsevier).

The model's practical output is a quantitative prediction: which pharmacological interventions, at which doses and timing, are most likely to prevent reward-memory consolidation in a given patient profile. That prediction can be packaged as a clinical decision-support tool. A clinician inputs patient characteristics, current medication, and stage of treatment; the model outputs a risk score and a suggested intervention strategy.

This is where the Africa Healthcare Innovation Fellowship becomes relevant. The CCT model was developed in silico, calibrated against published literature, and validated against pre-registered hypotheses. It has not yet been tested against real clinical data from African populations. The fellowship's field-based structure would allow me to pilot a CCT-based screening instrument in a Nigerian clinical setting, adapting the model's parameters to local patient demographics, medication availability, and healthcare workflows.

The innovation here is the translation of a validated computational model into a tool that a clinician with a smartphone and a basic internet connection can use. Nigeria has fewer than 5,000 psychiatrists for a population exceeding 200 million. Task-shifting to primary care is the only viable pathway for scaling addiction care. A computational screening tool that runs on existing hardware and requires no specialized training to interpret is a context-relevant solution.

The model is ready for field testing, not clinical deployment. The fellowship's mentorship component is essential for navigating the regulatory, ethical, and practical considerations of piloting a computational health tool in a Nigerian clinical setting. I have the technical skills, the pharmacological background, and the validated model. What I need is the field access and the guidance that AHIF provides.

PROJECT PROPOSAL

Project title: Field pilot of a CCT-based relapse risk screening tool for substance use disorder patients in Nigerian primary care settings.

Problem statement: Substance use disorders are under-addressed in Nigerian healthcare. The treatment gap is driven by a shortage of specialist mental health providers, limited access to pharmacological interventions, and the absence of validated, low-cost screening tools adapted to local contexts. The CCT model, a validated computational framework for reward-memory consolidation, offers a quantitative basis for relapse risk assessment, but it has not been tested against real clinical data.

Proposed solution: A 14-week field pilot during which I will adapt the CCT model into a screening instrument, test it with clinicians and patients at a partner clinic, and produce a feasibility report with preliminary validation data. The instrument will be a structured questionnaire plus a computational scoring engine that runs on a laptop or smartphone. No specialized hardware is required.

Activities:
- Weeks 1-2: Identify and secure a partner clinic in Nigeria; obtain ethical approval; finalize the screening instrument design based on the CCT model's three axes (dopaminergic sensitivity, NMDAR-related memory function, affective contrast).
- Weeks 3-8: Recruit 30 to 50 participants receiving treatment for substance use disorders; administer the screening instrument; collect baseline clinical data.
- Weeks 9-12: Analyze data against the CCT model's predictions; calibrate model parameters to local population characteristics; document discrepancies and adaptations.
- Weeks 13-14: Prepare a feasibility report, including preliminary validation statistics, implementation barriers, and a roadmap for a larger study.

Expected outcomes:
- A feasibility assessment of the CCT screening instrument in a Nigerian clinical setting.
- Preliminary data on whether the model's predictions correlate with clinician-assessed relapse risk.
- A documented, reproducible protocol for adapting computational pharmacology models to African clinical contexts.
- A clear go/no-go recommendation for a larger validation study.

Alignment with AHIF: The project is field-based, context-relevant, and designed for scalability. It addresses an under-recognized healthcare challenge in Africa. It leverages my dual expertise in pharmacology and computational modeling. It produces a tangible deliverable within the fellowship's 14-week timeframe. The mentorship component will be used to navigate ethical approval, clinical partnership, and data governance issues specific to Nigerian healthcare.

Risks and mitigation: Participant recruitment may be slower than expected; mitigation is a pre-identified backup clinic and a streamlined consent process. Ethical approval timelines may exceed the fellowship window; mitigation is early submission and engagement with a local ethics board that has experience with digital health research. The CCT model may require substantial adaptation to local data; this is an expected outcome, not a failure, and will be documented as part of the feasibility report.

I am seeking the fellowship's structure, mentorship, and field network to conduct this pilot, not funding for the model's development, which is complete. The model is ready. The clinical question is ready. The fellowship is the mechanism to connect them.

MOTIVATION LETTER

The CCT model, a tripartite pharmacological framework for preventing reward-memory encoding in addiction, addresses a healthcare gap that African health systems have not yet named as a priority. Substance use disorders across Nigeria and the continent remain under-diagnosed, under-treated, and under-funded, while the pharmacological mechanisms that drive relapse are well-characterized in the global literature. My work builds a computational bridge between those mechanisms and clinical practice. The Africa Healthcare Innovation Fellowship is the first programme I have found that explicitly rewards this kind of context-relevant, field-testable intervention rather than requiring a laboratory or a clinical trial infrastructure I do not have.

The model is a calibrated, tested system. I built a three-axis ODE model coupling dopaminergic reward prediction error, NMDAR-dependent long-term potentiation, and affective contrast, then calibrated all 14 free parameters using Bayesian MCMC with priors elicited from a systematic screen of 1,847 records. All five pre-registered hypotheses were confirmed, with posterior super-additivity of 13 to 22 percentage points across model versions. Three sole-authored preprints are under review at peer-reviewed journals. What the model now needs is a deployment pathway: a screening tool or clinical decision-support prototype that a Nigerian clinician can use to assess relapse risk and guide medication choice. That is precisely the hands-on, field-based work AHIF supports.

My background fits the fellowship's interdisciplinary mandate. I hold a B.Pharm from the University of Ibadan with a German equivalent grade of 1.9, I am a licensed pharmacist, and I am enrolled in the M.Sc. Digital Health programme at the Hasso Plattner Institute in Potsdam. I have built and validated computational pipelines across addiction neuroscience, protein machine learning, and dynamical systems. I have also worked as a clinical pharmacist and as a research assistant on antimicrobial resistance genomics. I know how to read a prescription and how to write a simulation. Few applicants to this fellowship will bring both.

What I need from AHIF is mentorship and a field network to translate a validated computational model into a tool that works in a Nigerian clinic. The 14-week field-based structure is the correct format for this: I need to sit with clinicians, understand their workflow, and adapt the CCT screening instrument to real patient data. The fellowship's emphasis on scalable, context-relevant solutions matches the CCT model's design, which runs on a laptop and requires no specialized hardware.

I am applying to AHIF because it is the only programme I have identified that combines mentorship, field placement, and a focus on African healthcare innovation without requiring me to abandon my computational methods or relocate permanently. The model is ready. The clinical context is ready. The fellowship is the missing link.

RESEARCH STATEMENT

The Conjunctive Consolidation Threshold (CCT) model is a computational framework for understanding why some individuals transition from drug use to addiction and, more importantly, how that transition can be pharmacologically interrupted. The core thesis is that reward-memory consolidation requires the simultaneous activation of three coupled systems: dopaminergic reward prediction error signaling, NMDAR-dependent long-term potentiation in reward circuitry, and affective contrast between drug and non-drug states. Interrupting any one axis is insufficient; the threshold for consolidation is conjunctive.

I operationalized this thesis as a system of ordinary differential equations with three coupled axes, solved using RK45 integration. The model has 14 free parameters, all calibrated using Bayesian MCMC with the DEMetropolisZ sampler in PyMC. Priors were elicited from a systematic literature screen of 1,847 records covering dopaminergic signaling, NMDAR pharmacology, and affective neuroscience. The model was pre-registered with five hypotheses before calibration. All five were confirmed. Posterior analysis showed super-additive effects of multi-axis intervention, ranging from 13 to 22 percentage points over single-axis approaches across model versions. Three sole-authored preprints describing the model, its calibration, and its pharmacological implications are currently under review at peer-reviewed journals. A co-authored paper is under review at Alcohol (Elsevier).

The model's practical output is a quantitative prediction: which pharmacological interventions, at which doses and timing, are most likely to prevent reward-memory consolidation in a given patient profile. That prediction can be packaged as a clinical decision-support tool. A clinician inputs patient characteristics, current medication, and stage of treatment; the model outputs a risk score and a suggested intervention strategy.

This is where the Africa Healthcare Innovation Fellowship becomes relevant. The CCT model was developed in silico, calibrated against published literature, and validated against pre-registered hypotheses. It has not yet been tested against real clinical data from African populations. The fellowship's field-based structure would allow me to pilot a CCT-based screening instrument in a Nigerian clinical setting, adapting the model's parameters to local patient demographics, medication availability, and healthcare workflows.

The innovation here is the translation of a validated computational model into a tool that a clinician with a smartphone and a basic internet connection can use. Nigeria has fewer than 5,000 psychiatrists for a population exceeding 200 million. Task-shifting to primary care is the only viable pathway for scaling addiction care. A computational screening tool that runs on existing hardware and requires no specialized training to interpret is a context-relevant solution.

The model is ready for field testing, not clinical deployment. The fellowship's mentorship component is essential for navigating the regulatory, ethical, and practical considerations of piloting a computational health tool in a Nigerian clinical setting. I have the technical skills, the pharmacological background, and the validated model. What I need is the field access and the guidance that AHIF provides.

PROJECT PROPOSAL

Project title: Field pilot of a CCT-based relapse risk screening tool for substance use disorder patients in Nigerian primary care settings.

Problem statement: Substance use disorders are under-addressed in Nigerian healthcare. The treatment gap is driven by a shortage of specialist mental health providers, limited access to pharmacological interventions, and the absence of validated, low-cost screening tools adapted to local contexts. The CCT model, a validated computational framework for reward-memory consolidation, offers a quantitative basis for relapse risk assessment, but it has not been tested against real clinical data.

Proposed solution: A 14-week field pilot during which I will adapt the CCT model into a screening instrument, test it with clinicians and patients at a partner clinic, and produce a feasibility report with preliminary validation data. The instrument will be a structured questionnaire plus a computational scoring engine that runs on a laptop or smartphone. No specialized hardware is required.

Activities:
- Weeks 1-2: Identify and secure a partner clinic in Nigeria; obtain ethical approval; finalize the screening instrument design based on the CCT model's three axes (dopaminergic sensitivity, NMDAR-related memory function, affective contrast).
- Weeks 3-8: Recruit 30 to 50 participants receiving treatment for substance use disorders; administer the screening instrument; collect baseline clinical data.
- Weeks 9-12: Analyze data against the CCT model's predictions; calibrate model parameters to local population characteristics; document discrepancies and adaptations.
- Weeks 13-14: Prepare a feasibility report, including preliminary validation statistics, implementation barriers, and a roadmap for a larger study.

Expected outcomes:
- A feasibility assessment of the CCT screening instrument in a Nigerian clinical setting.
- Preliminary data on whether the model's predictions correlate with clinician-assessed relapse risk.
- A documented, reproducible protocol for adapting computational pharmacology models to African clinical contexts.
- A clear go/no-go recommendation for a larger validation study.

Alignment with AHIF: The project is field-based, context-relevant, and designed for scalability. It addresses an under-recognized healthcare challenge in Africa. It leverages my dual expertise in pharmacology and computational modeling. It produces a tangible deliverable within the fellowship's 14-week timeframe. The mentorship component will be used to navigate ethical approval, clinical partnership, and data governance issues specific to Nigerian healthcare.

Risks and mitigation: Participant recruitment may be slower than expected; mitigation is a pre-identified backup clinic and a streamlined consent process. Ethical approval timelines may exceed the fellowship window; mitigation is early submission and engagement with a local ethics board that has experience with digital health research. The CCT model may require substantial adaptation to local data; this is an expected outcome, not a failure, and will be documented as part of the feasibility report.

I am seeking the fellowship's structure, mentorship, and field network to conduct this pilot, not funding for the model's development, which is complete. The model is ready. The clinical question is ready. The fellowship is the mechanism to connect them.

AI Draft — Africa Healthcare Innovation Fellowship (AHIF)