Results

7.7h

saved per week for doctors.

80%

reduction in documentation workload.

2.5x

times more detailed reports compared to manual entries.

The complexity of medical terminology


Developing a highly accurate language model for medical terminology was particularly challenging due to the complexity and specificity of medical terms.

Real-time functionality for seamless workflows

It was essential to ensure the solution operates in real time to maintain the uninterrupted workflow of practicing physicians.

Secure handling of sensitive patient data

The processing of sensitive patient data also required functioning solutions in order to be able to guarantee the higher requirements in this area at all times.

Our approach

Data-driven innovation: Our data-based solution was developed to significantly increase efficiency in medical documentation.

AI for healthcare: With artificial intelligence, we have automated medical documentation and thus created valuable time for doctors to provide direct patient care.

Language technology for medicine: Our cutting-edge speech technology accurately transcribes and structures medical conversations, ensuring comprehensive and precise documentation.

Industry perspective

30%

of German healthcare CEOs are already leveraging AI solutions to optimise workflows and automate administrative tasks, freeing up more time for patient care.

$188bil.

is the projected global market value by 2030 of AI in healthcare, which offers immense potential for innovation and efficiency improvements.

Overview

Project scope

The project included the development of an AI-based solution for automated transcription and documentation of medical conversations. The focus was on the precise recognition of medical terms, the preparation of structured reports and integration into existing practice management systems.

Key objectives & deliverables

1

Development of a precise LLM for the recognition of medical terminology.

2

Create a robust natural language processing (NLP) model to extract relevant information from medical conversations

3

Design a user-friendly interface for optimal use.

4

Ensuring data protection and security.

Adiu health is an innovative start-up that specialises in automating medical documentation. With the help of AI, Adiu is revolutionising the way doctors document their patient treatments through tailor-made summaries.

Health & AI

The healthcare sector is facing a digital transformation in which artificial intelligence is playing an increasingly important role. AI applications have the potential to increase process efficiency, improve the accuracy of diagnoses, and personalise patient care. From medical image processing to drug development, AI opens up new opportunities to meet healthcare challenges.

Lorem ipsum

Every second doctor experiences burnout — the main causes: too much paperwork, bureaucracy and too long working hours.

Medscape Physician Burnout and Depression Report, 2024

Projektphasen

Stage 1
Preparation & prototyping

Defining the capabilities of the application: Automated transcription and documentation of medical conversations

Implementation of a minimal prototype: Demonstration of options for transcribing and summarising doctor-patient conversations

Designing an intuitive front end

Stage 2
Implementation & deployment

Deploy to AWS: Hosting a complete solution on AWS

Implementation of an intuitive front end: Building a modern React-based web front end

Stage 3
Evaluation & Verbesserung

Evaluation: Using Human Feedback to the Chatbot Results and the Overall Solution for Enhancements

Verbesserung: Refining the Chatbot Responses and the Frontend for Rollout to External Customers

Stage 4
handover

Comprehensive training and knowledge transfer: Documentation and implementation of training courses to ensure seamless takeover and independent registration by the customer.

Long-term partnership: Equip Adiu with the tools and support for continuous improvement while ensuring that Perelyn's expertise is always available.

Phase 1
Vorbereitung & Prototyping

Definition der Fähigkeiten der Anwendung: Automatisierte Transkription und Dokumentation medizinischer Gespräche

Implementierung eines minimalen Prototyps: Demonstration der Möglichkeiten der Transcription und Zusammenfassung von Doktor-Patienten Gesprächen

Design eines intuitiven Frontends

Phase 2
Implementierung & Deployment

Bereitstellen auf AWS: Hosting einer vollständigen Lösung auf AWS

Implementierung eines intuitiven Frontends: Aufbau eines modernen React-basierten Web-Frontends

Internal productionalisation: rollout to different departments within the company​

Phase 3
Evaluation & Verbesserung

Evaluation: Intensive System und Funktionstests mit kleiner Gruppe von Domain Experten (Ärzten)

Verbesserung: Integration und Umsetzung des Feedbacks aus der Testgruppe in Bezug auf Funktionsumfang und Benutzung

Phase 4
Übergabe

Umfassende Schulung und Wissenstransfer: Dokumentation und Durchführung von Schulungen zur Gewährleistung einer nahtlosen Übernahme und der selbständigen Registrierung durch den Kunden.

Langfristige Partnerschaft: Adiu mit den Werkzeugen und der Unterstützung für kontinuierliche Verbesserung ausstatten und gleichzeitig sicherstellen, dass Perelyns Expertise jederzeit verfügbar ist.

Which solution are you looking for?

Overview

We have developed an innovative solution that revolutionises medical documentation.
Our software leverages cutting-edge AI technologies to enable precise and efficient automatic transcription of medical conversations — fully compliant with GDPR!

Business benefits

The AI-based transcription solution optimizes documentation processes across industries — from healthcare and law to education and finance to media and customer service. It enables efficient protocols, archiving and customer interactions and adapts flexibly to individual requirements.

Technical architecture & integration

Our platform is modular and can be adapted to individual requirements. By using cloud-based technologies, we enable high scalability and can therefore also process large amounts of data.

Adiu

Consultation creation

Transcription

Summarisation

Crystal clear web & mobile audio recording

Medical history processing using previous doctors’ insights

Thorough patient analysis

Blazingly fast NVIDIA A10 Tensor Core GPU

Near real time batch processing on EU hosted AWS EC2 machines

Extremely accurate OpenAI Whisper transcriptions

State of the art LLM by Anthropic -Claude 3.5 Sonnet

Serverless computing utilising AWS Lambda

Detailed medical documentation

Key tech

Transcription

AWS Bedrock, Amazon Comprehend, Google VertexAI, Azure Open AI, HuggingFace Transformers, LangChain, LangSmith, LLamaIndex

Summarisation

Anthropic, OpenAI, AWS Bedrock

Modern Frontend

React

Data security

AWS Private VPC, all processing and storage is done in the EU.

Programming languages

AWS, ECS, CloudFront, MongoDB

Programming languages

Python, TypeScript

Ongoing process

Building Partnerships for Innovation

Adiu Health leverages AI to transform vast amounts of medical guideline literature into knowledge graphs. For the efficient and secure storage and utilisation of these knowledge graphs, Adiu Health relies on Neo4j’s state-of-the-art graph database solution.

By integrating these knowledge graphs with LLMs, Adiu Health enables physicians to perform precise and personalised searches or enhance consultation summaries with automated follow-up checks, such as guideline-based reviews of potential medications or treatment strategies.

Relevant insights

Generative AI
The next big leap in retail

Discover the impact of generative AI in retail. This technology unites departments, personalises content, and transforms customer experiences. Leading brands such as Coca-Cola and Walmart are already using their potential to optimise operations and drive innovation. Explore the future of retail with generative AI...

#LLM
#Generative AI
Sebastian Fetz
28.2.2024
Read
Generative AI
How do you use LLMs with internal company data?

This article explores how large language models such as ChatGPT can be integrated with proprietary data. The recent advent of large language models opens up numerous opportunities for companies, but connecting with proprietary data is a challenge in such a transformation. This article discusses the various approaches to solve this problem, such as fine-tuning and contextual learning. In addition, the associated challenges and risks are addressed.

#LLM
#Generative AI
Dominik Filipiak
19.2.2024
Read
Generative AI
Superhuman Assistant - LLMs as a Strategic Priority

In this article, our CEO Sebastian highlights the transformative power of large language models (LLMs) in a corporate context and provides guidance for their implementation. This is not just a dream of the future, but a strategic necessity that future-oriented companies are taking on today to drive groundbreaking innovations.

#LLM
#AI Strategy
#Generative AI
#LLMOps
Sebastian Fetz
10.1.2024
Read

Contact us

Name*
Email*
message*
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Help us improve our website. By “Accept” click, do you agree that we store a cookie on your device to analyse the use of the website. Read our Privacy statement for more information.
Something went wrong. Please try again.