Evidence & innovation portfolio
Explore public research, innovation, policy, and implementation records. Project stage, product status, relationship, contributors, institutions, and sources remain distinct and inspectable.
The public portfolio includes only records approved for disclosure; confidential or unverified work is withheld.
Canonical scientific records
Study aim, preprint provenance, active journal stage, Network-derived author relationship, manuscript affiliations, controlled tags and verification date remain separate and inspectable.
A selection of our work. Some projects in progress or peer review are not disclosed under confidentiality agreements with their investigators.
A percentage describes a proportion within the stated group. A risk ratio compares two risks: 1 means equal risks; 2 means twice the risk in the stated comparison, not twice the number of patients. Association alone does not establish causation.
The estimated number of people treated for one additional benefit over a specified period and comparator. It depends on baseline risk and uncertainty.
Read the BMJ methods article ↗A quality-adjusted life-year (QALY) combines length and quality of life. Net monetary benefit (NMB) values health gains at the study’s stated threshold, then subtracts costs. A cost-effectiveness probability summarizes model uncertainty; it is not the probability that a patient improves.
Read about cost-effectiveness uncertainty ↗Read the group order, units, study design and follow-up alongside each estimate. Counts remain whole numbers; monetary and percentage estimates use two decimal places where supported. See the source for confidence intervals and limitations.
Pediatric trauma and reconstructive surgery
This case report describes the staged orthoplastic reconstruction of an 8-cm open Achilles tendon defect following crush trauma in a 7-year-old child, focusing on the reconstructive strategy used to restore tendon continuity and soft-tissue coverage after severe injury.
Retinopathy of prematurity and health economics
This study evaluates the cost-effectiveness of topical ketorolac and nepafenac prophylaxis for early-stage retinopathy of prematurity using multicenter real-world clinical and economic data from Colombia.
Incremental value expressed as net benefit at the study willingness-to-pay threshold.
Share of probabilistic simulations favoring the prophylaxis strategy at the study threshold.
Retinopathy of prematurity
This study evaluates real-world clinical outcomes associated with topical ketorolac and nepafenac prophylaxis for early-stage retinopathy of prematurity across a multicenter care network in Colombia.
Observed progression in prophylaxis and comparison groups, respectively.
Estimated number treated to prevent one additional Stage 3-or-higher outcome.
Observed mean time to remission in prophylaxis and comparison groups.
Transplantation and health-system access
This study evaluates graft-specific geographic accessibility to transplant-capable services across Colombia and identifies populations living beyond clinically relevant travel-time thresholds to support transplant-network planning and identify underserved areas.
Retinopathy of prematurity and health economics
This study compares the cost-effectiveness of bevacizumab and ranibizumab for treatment-requiring retinopathy of prematurity using multicenter real-world clinical and economic data.
Observed retreatment after bevacizumab and ranibizumab, respectively.
Incremental net value favoring bevacizumab at the study threshold.
Share of probabilistic simulations favoring bevacizumab.
Critical care access and health-system planning
This study evaluates population-level travel-time access to intensive-care services across Colombia and identifies areas with prolonged geographic access to inform critical-care policy, service planning, and network design.
Retinopathy of prematurity workforce and access
This study characterizes the distribution of the retinopathy of prematurity workforce in Colombia, evaluates geographic accessibility to care, and examines emerging practice patterns to identify gaps relevant to national service planning.
Obstetric critical care
This study evaluates rural–urban differences in obstetric intensive-care admission, organ support, resource use, and maternal–fetal outcomes within a Colombian hospital cohort from 2014 through 2018.
Health economics and decision science
This methodological study evaluates the empirical stability of the incremental cost-effectiveness ratio relative to net monetary benefit under probabilistic uncertainty, examining which metric communicates decision uncertainty more reliably.
Artificial intelligence and biomedical engineering
This development program is defining a governed pathway for retinal-image acquisition, algorithm evaluation, specialist review, and referral support in retinopathy of prematurity without making unverified clinical-performance claims.
Technology evidence · human-led by design
This evidence family documents concepts, physical prototypes, task-specific development benchmarks, validation gaps and human decision boundaries. microscopIA develops technology for settings where specialist capacity is limited without presenting investigational systems as clinical products.
microscopIA is developing an integrated research system for retinal imaging in preterm infants. The device is intended to support a high-resolution camera, primarily a compatible smartphone, to improve inspection and capture of infant retinal images.
Research models analyse de-identified retinal images to study image quality and features of retinopathy of prematurity (ROP). The intended goal is to help identify infants with concerning findings or higher risk for referral to a ROP-trained ophthalmologist, who retains diagnosis and treatment decisions. Future use is being evaluated for appropriately trained neonatologists, neonatal nurses and physicians, including general practitioners where permitted, working under approved institutional protocols and local scope-of-practice rules.
Ungradable image or failed acquisition → repeat or escalate under the approved protocol. There is no AI-created ‘no referral’ pathway.
Latest supplied AI development record
Source last modified 19 November 2025. Values are shown by model task because the supplied record does not establish performance of an integrated device–AI system.
Stage-related image classification
92.03Accuracy · reported scoreGCViT · 1,099 public-dataset imagesPlus-disease classification
93.10Accuracy · reported scoreResNet50 · public development datasetVessel segmentation
76.81%Intersection over UnionU-Net · HVDROPDBThe stage and plus values are reproduced as reported; the source does not print a percent sign for them. These are separate public-dataset development tasks with incomplete evaluation details. They are not directly comparable, externally or clinically validated, or evidence of device-level performance. No specificity, AUROC or confidence intervals are claimed.
The public video documents a physical microscopy prototype. The programme investigates automated image acquisition and software-assisted flagging of candidate fields, quality issues or samples for review by trained laboratory professionals. It does not confirm or exclude malaria, establish species or parasite density, choose treatment, or replace laboratory quality assurance and clinical judgement.
Any future application to another microscopy-based tropical disease requires its own intended use, dataset, validation plan and regulatory assessment.
Research priorities start with settings where specialist people and infrastructure are scarce.
Technology may support acquisition and prioritisation; qualified professionals remain responsible for decisions.
Safety engineering, usability, governance, training and independent validation precede clinical use.
Collaboration areas: ROP and pediatric ophthalmology, neonatology, optics and imaging, human factors, device engineering, clinical-AI validation, laboratory microscopy, quality systems and regulation.
Explore responsible fundingNetwork intelligence
One canonical registry brings together articles, preprints, chapters, guidelines, conference work, patents, software, and other outputs. Relationship tags distinguish microscopIA affiliation, collaboration, and independent member scholarship.
Every public output requires a verified member identity, an inspectable source, and human attribution review. Google Scholar supports discovery but is never the sole proof.
Checking member profiles and scholarship
Turn work into evidence
Strong collaboration begins with a worthwhile question, governed data, and a clear decision audience.
