Upload README.md with huggingface_hub

README.md

@@ -100,11 +100,10 @@ epidemiology. The dataset is designed for cross-sectional grading: predicting
 DR stage from a fundus image, optionally conditioned on the limited information
 available at the point of grading. We document the construction, the grounding
 literature, and a reproducible **distribution audit** that validates the
-synthetic context against its source effect sizes. We are explicit about two
-boundaries: the images are not African, and the clinical context is synthetic
-and not patient-linked. The dataset is therefore a scaffold for building and
-benchmarking grading pipelines and for teaching — **not** a source of clinical
-or epidemiological findings.
+synthetic context against its source effect sizes. Two boundaries should be
+clear: the images are not African, and the clinical context is synthetic and not
+patient-linked. The dataset is meant for building and testing grading pipelines
+and for teaching, not as a source of clinical or epidemiological findings.
 
 ---
 
@@ -120,18 +119,17 @@ hardware, and care-seeking behaviour. When such models are evaluated on African
 populations they frequently degrade, a manifestation of dataset shift that is
 both a scientific and an equity problem.
 
-The **DR-Africa** family is an attempt to build infrastructure that takes this
-gap seriously. Its first objective is a *multimodal* resource: DR research is
-moving beyond the image alone toward models that combine retinal imaging with
-systemic clinical variables (glycaemic control, diabetes duration, blood
-pressure, renal function), because those variables are the true drivers of
+The **DR-Africa** family addresses this gap. Its first objective is a
+*multimodal* resource: DR research is moving beyond the image alone toward models
+that combine retinal imaging with systemic clinical variables (glycaemic control,
+diabetes duration, blood pressure, renal function), which are the main drivers of
 incidence and progression. Real linked image–EHR cohorts from Africa are scarce,
-so we bootstrap the pipeline with **epidemiology-grounded synthetic** clinical
-data: synthetic records whose marginal distributions and conditional structure
-are parameterised from published SSA cohort studies and meta-analyses.
+so we start with **epidemiology-grounded synthetic** clinical data: synthetic
+records whose marginal distributions and conditional structure are taken from
+published SSA cohort studies and meta-analyses.
 
 This repository, **DR-Grading**, is the entry point of the family and the
-narrowest in scope by design (separation of concerns). It exposes only the
+narrowest in scope (separation of concerns). It exposes only the
 inputs realistically available *at the moment of grading* — the image plus three
 context fields — and the grade label. The richer systemic record lives in
 [`dr-progression`](../dr-progression); screening-prevalence evaluation lives in
@@ -155,8 +153,8 @@ proliferative DR (PDR). These map to the five source folders used here
 
 **Existing DR image datasets.** EyePACS, APTOS 2019, Messidor / Messidor-2,
 IDRiD, DDR, and DeepDRiD are the most widely used. They differ in grading
-protocol, image quality, and population, and — crucially for this work — none
-ships paired structured clinical data, and none is African in origin.
+protocol, image quality, and population. None ships paired structured clinical
+data, and none is African in origin.
 
 **Epidemiology of DR in Africa.** Pooled SSA estimates place any-DR prevalence
 around 30–35% of people with diabetes and PDR at ~1% in population-based studies
@@ -323,8 +321,8 @@ about real African patients.
    property of the generator, not evidence about patients.
 3. **Referral-skewed labels.** The class balance reflects a clinic/referral
    population, not screening; uncorrected accuracy overstates screening utility.
-4. **Minimal context by design.** Labs and comorbidities are intentionally absent
-   (they belong to `dr-progression`).
+4. **Minimal context.** Labs and comorbidities are not included here (they belong
+   to `dr-progression`).
 
 ---
 
@@ -332,15 +330,14 @@ about real African patients.
 
 - **Synthetic data are flagged.** Every record carries `synthetic = True`. No
   attempt should be made to interpret a synthetic record as a real individual.
-- **Representation honesty.** Calling a dataset "Africa-grounded" while the
-  imagery is non-African risks misrepresentation; we state the provenance
-  prominently and treat the images strictly as a transfer-learning source.
-- **Data sovereignty.** The roadmap toward real African imaging and linked
-  records must be governed by local consent, ethics approval, and data-sovereignty
-  principles; this synthetic scaffold is explicitly a placeholder, not a substitute.
-- **Intended benefit.** The resource exists to *amplify* DR research relevant to
-  Africa by lowering the barrier to multimodal experimentation, not to replace
-  real-world data collection.
+- **Representation.** Calling a dataset "Africa-grounded" while the imagery is
+  non-African could mislead, so we state the provenance plainly and treat the
+  images only as a transfer-learning source.
+- **Data sovereignty.** The move toward real African imaging and linked records
+  must follow local consent, ethics approval, and data-sovereignty principles.
+  This synthetic dataset is a placeholder, not a substitute.
+- **Purpose.** The dataset exists to support DR research relevant to Africa by
+  making multimodal work easier to start, not to replace real-world data collection.
 
 ---