About the client
Based in the United States, the client develops enterprise software for healthcare providers. Its EMR platform helps hospitals and clinics manage patient records, coordinate care, and meet strict regulatory requirements. Over the years, the system became a core tool for day-to-day operations across multiple facilities, reducing administrative effort and supporting more consistent clinical workflows.
Country:
USA
Industry:
Healthcare software
Service:
AI integration for EMR/EHR systems
Duration
2023 – ongoing
Model:
B2B
Team size:
1 Business Analyst, 2 AI/ML Engineers, 2 Backend Developers, 1 DevOps Engineer, 1 Front-end Developer
Fast and reliable data handling is critical in healthcare systems as clinical records grow in volume and complexity. Manual reporting and review still take time and often lead to uneven results, even in advanced EMR systems. For vendors serving hospitals and clinics, that means higher costs and slower adoption.
Beyond being a global trend, AI offered real practical value for the company. It could take over repetitive yet critical tasks, such as generating reports, reviewing data for inconsistencies, and assisting clinicians with quick, data-based decisions. The goal was to reduce manual effort and improve the system’s accuracy and usability in real clinical settings.
Client’s challenge
The company needed engineering support to introduce AI into its product. Building a reliable model and integrating it into a regulated EMR environment required specific skills in data science, backend architecture, and compliance. Recruiting such specialists locally proved difficult and time-consuming. To meet release deadlines and control costs, the company decided to bring in an external engineering partner.
- No in-house expertise in applied AI
- Patient data stored in both forms and free-text notes
- Heavy manual effort in preparing and reviewing reports
- New AI features needed to work inside existing clinician workflows
- Need to maintain HIPAA / HITECH compliance
Solution
Neklo’s team extended the client’s EMR platform with AI features for automating reports and assisting in clinical data review.
Architecture and stack
We used PyTorch to design and train the AI models behind report automation and data analysis. After training, we deployed the models on Vertex AI within the client’s Google Cloud environment, where they automatically adjust the amount of processing power based on load.
Data flow and integration
Integrating AI into an existing EMR was not a plug-and-play task. Data formats differed across modules, and some were not suitable for direct model input. Our engineers tested several options for preparing and routing the data — from using intermediate buffers to direct API calls — before choosing a direct, API-based setup with in-memory preprocessing. This approach was fast enough for executing real-time reporting and fit into the EMR’s existing security and compliance environment. They also adjusted the preprocessing logic to handle uncommon or irregular data patterns in clinical notes, like typos in doctors’ notes or terminology misuse.
Compliance and security
We also made sure the new AI functions followed HIPAA and HITECH rules. The automation worked entirely inside the EMR’s secure environment: the module processed de-identified data through encrypted APIs; each operation used role-based access control linked to EMR user permissions; all AI requests were logged for audits, including user, time, and data type; no patient information was stored or cached outside the EMR.
User interface integration
On the front end, we added new functionality to the existing React interface, enabling clinicians to create AI-assisted reports and view alerts without needing to open another app, switch tabs, or change their workflow.
Show all challenges
The process
Having a solid expertise in AI solutions for healthcare, we knew exactly how to address the client’s business goals.
1. Requirements Alignment
A cross-functional team of AI engineers, backend developers, DevOps specialists, and a business analyst worked with the client’s stakeholders to review existing workflows, map data sources, and define integration points within the EMR platform.
2. Prototyping and Validation
Once the scope and requirements were agreed, we built a proof of concept to test several machine learning approaches. The goal was to ensure the model could handle the type, quality, and volume of the client’s data and reach the level of accuracy needed for real use, for example, detecting unexpected patterns without producing false positives. The prototype helped verify data consistency, tune preprocessing logic, and validate model outputs before production rollout.
3. Project Management Routine
Throughout the engagement, the teams worked in shared Jira and Confluence spaces, held daily check-ins via Slack, and tracked progress against delivery milestones. We ran regular demo sessions for the client to review intermediate results and clarify expectations around data handling and report outputs. This helped refine the solution as the model was tested on real clinical data and keep the project on schedule.
1. Discussing your needs
1. Discussing your needs
2. Coordination
Here, we define team structure and allocate resources. We also outline and set up communication channels and progress-tracking tools so you have a transparent view of your project at all times. Planning is the key to efficient development and building long-term relationships with our clients.
3. Onboarding
Once everyone agrees on the plan, we gather the tools, environments, data, and 3rd party integrations we need to keep development running smoothly. Our testing environments let you see your project in action and allow us to identify any potential issues early in the process so solutions are built securely and efficiently from day one.
4. Development
Being a professional Hyvä themes developer requires more than development skills. Seamless deployments with minimal downtime, backups, data security, conversion, and SEO are taken into account from the first line of code, along with the visual design and user experience. Transparent progress reporting and regular communications ensure you always know the status of your project.
5. Training & Documentation
Our projects aren’t complete until our clients are comfortable using them. We provide training and ongoing support so your team gets the maximum benefit of our work. And our documentation of the entire process makes it easy to adapt the project to your changing needs over time.
Technologies
Front-end
Material UI
React
Redux-Toolkit
Back-end
Cloud SQL
Elasticsearch
FastAPI
GraphQL
Python
Redis
AI/Data
Anomaly Detection
In-memory
Models
Preprocessing
PyTorch
Vertex AI
DevOps / Infrastructure
Cloud Functions
Cloud Healthcare API
.svg)
Control (RBAC)
Docker
End-to-End Encryption
Google Cloud Platform
Kubernetes (GKE)
Role-Based Access
Deliverables
AI-assisted reporting feature
Clinical data preprocessing
Model training and evaluation
Model deployment in the client’s environment
Secure EMR data integration via APIs
Role-based access control for AI features
Audit logging for compliance
UI updates to support in-workflow use
Result
Introducing machine learning and AI-driven reporting improved both the speed and clarity of clinical documentation. Clinicians saved an average of 13 minutes per patient, which, in a typical outpatient schedule, frees several hours per day that would otherwise be spent on manual record updates. Documentation became more consistent, reducing the need for clarifications and follow-up edits.
With less time spent entering data during visits, clinicians had more attention available for conversation and examination. In patient surveys, 93% reported that care providers felt more present and engaged. Overall workflow efficiency increased by 24%, helping clinics handle daily caseloads with less time pressure.
13
min
Saved per visit
+65
%
Documentation quality
93
%
Patient engagement rating
+24
%
Encounter efficiency per hour
.png)
