LG AI Research Develops AI Model on AWS for Rapid Cancer Diagnosis
LG AI Research has developed EXAONEPath, a new pathology foundation model for cancer diagnosis, using AWS cloud infrastructure. The AI model can analyze microscopic tissue samples with 86.1% accuracy across six benchmarks, reducing genetic testing time from two weeks to under one minute. Using AWS services, including Amazon SageMaker and NVIDIA GPUs, LG AI Research trained the model on 285 million data points and 35,000 tissue samples, cutting data management costs by 35% and data preparation time by 95%. The model is part of EXAONE, LG's 300-billion-parameter multimodal foundation model.
LG AI Research ha sviluppato EXAONEPath, un nuovo modello fondazionale per la patologia dedicato alla diagnosi del cancro, utilizzando l'infrastruttura cloud di AWS. Il modello AI può analizzare campioni di tessuto microscopici con un'accuratezza del 86,1% su sei benchmark, riducendo il tempo di test genetico da due settimane a meno di un minuto. Utilizzando i servizi AWS, tra cui Amazon SageMaker e GPU NVIDIA, LG AI Research ha addestrato il modello su 285 milioni di punti dati e 35.000 campioni di tessuto, riducendo i costi di gestione dei dati del 35% e il tempo di preparazione dei dati del 95%. Il modello è parte di EXAONE, il modello fondazionale multimodale di LG con 300 miliardi di parametri.
LG AI Research ha desarrollado EXAONEPath, un nuevo modelo fundamental para patologías en el diagnóstico del cáncer, utilizando la infraestructura en la nube de AWS. El modelo de IA puede analizar muestras de tejido microscópicas con una exactitud del 86.1% en seis indicadores, reduciendo el tiempo de prueba genética de dos semanas a menos de un minuto. Utilizando servicios de AWS, incluyendo Amazon SageMaker y GPU de NVIDIA, LG AI Research entrenó el modelo con 285 millones de puntos de datos y 35,000 muestras de tejido, reduciendo los costos de gestión de datos en un 35% y el tiempo de preparación de datos en un 95%. El modelo es parte de EXAONE, el modelo fundamental multimodal de LG con 300 mil millones de parámetros.
LG AI Research는 EXAONEPath를 개발했습니다. 이는 암 진단을 위한 새로운 병리 기초 모델로, AWS 클라우드 인프라를 사용합니다. AI 모델은 여섯 가지 기준에서 86.1%의 정확도로 미세 조직 샘플을 분석할 수 있으며, 유전자 검사 시간을 2주에서 1분 이내로 단축합니다. Amazon SageMaker 및 NVIDIA GPU를 포함한 AWS 서비스를 사용하여, LG AI Research는 2억8500만 데이터 포인트와 3만5000개 조직 샘플로 모델을 훈련시켜 데이터 관리 비용을 35% 절감하고 데이터 준비 시간을 95% 단축했습니다. 이 모델은 3000억 개의 매개변수를 가진 LG의 EXAONE 멀티모달 기초 모델의 일부입니다.
LG AI Research a développé EXAONEPath, un nouveau modèle de base en pathologie pour le diagnostic du cancer, utilisant l'infrastructure cloud d'AWS. Le modèle d'IA peut analyser des échantillons de tissus microscopiques avec une précision de 86,1% sur six références, réduisant le temps de test génétique de deux semaines à moins d'une minute. En utilisant les services AWS, y compris Amazon SageMaker et les GPU NVIDIA, LG AI Research a formé le modèle sur 285 millions de points de données et 35 000 échantillons de tissus, réduisant les coûts de gestion des données de 35 % et le temps de préparation des données de 95 %. Le modèle fait partie d'EXAONE, le modèle fondamental multimodal de LG comprenant 300 milliards de paramètres.
LG AI Research hat EXAONEPath entwickelt, ein neues fundamentales Modell für die Pathologie zur Krebsdiagnose, das die AWS-Cloud-Infrastruktur nutzt. Das KI-Modell kann mikroskopische Gewebeproben mit einer Genauigkeit von 86,1% über sechs Benchmarks analysieren und reduziert die genetische Testzeit von zwei Wochen auf weniger als eine Minute. Mit AWS-Diensten wie Amazon SageMaker und NVIDIA-GPUs hat LG AI Research das Modell mit 285 Millionen Datenpunkten und 35.000 Gewebeproben trainiert und die Datenverwaltungskosten um 35% sowie die Datenvorbereitungszeit um 95% gesenkt. Das Modell ist Teil von EXAONE, Lgs multimodalen Grundmodell mit 300 Milliarden Parametern.
- Reduces genetic testing time from 2 weeks to under 1 minute
- Achieves 86.1% accuracy in classifying cellular-level features
- Cuts data management and infrastructure costs by 35%
- Reduces data preparation time by 95%
- Shortens model training time from 60 days to one week
- None.
Insights
The development of EXAONEPath represents a significant technological breakthrough in cancer diagnostics. The 86.1% accuracy rate across six benchmarks is particularly impressive given the smaller dataset used compared to competitors. The reduction in genetic testing time from two weeks to under one minute could revolutionize cancer diagnosis workflows and potentially save countless lives through earlier intervention.
The model's ability to analyze microscopic tissue samples with high accuracy while maintaining speed demonstrates a remarkable balance of precision and efficiency. The planned expansion to detect additional cancer types suggests strong scalability potential. The 35% cost reduction in data management and 95% reduction in data preparation time indicate significant operational benefits for healthcare providers.
AWS's infrastructure capabilities shine in this implementation, particularly in handling the massive computational requirements for the 300-billion-parameter model. The reduction in model training time from 60 days to one week demonstrates exceptional optimization. The combination of Amazon SageMaker, S3 and FSx for Lustre creates a powerful tech stack that enables sub-millisecond latencies and hundreds of gigabytes per second throughput.
The ability to transfer terabytes of data in under an hour while processing 285 million data points and 35,000 high-resolution images showcases AWS's enterprise-grade capabilities in AI/ML workloads. This partnership strengthens AWS's position in the healthcare AI market.
LG Group’s AI think tank uses AWS to identify cancer risks earlier
Amazon SageMaker helps LG AI Research reduce genetic testing time from two weeks to less than one minute to accelerate patient diagnosis
EXAONEPath achieves an average accuracy of
“AWS allows us to accelerate our AI research, bringing accessible and rapid cancer screening closer to reality,” said Hwayoung (Edward) Lee, vice president of LG AI Research. “By leveraging AWS, we can train our pathology model on a vast dataset faster—securely, and cost-effectively. This enhances EXAONEPath’s processing capabilities for delivering personalized, efficient cancer treatments to improve patient outcomes. EXAONEPath has the potential to transform cancer diagnosis and treatment globally.”
Leveraging Amazon SageMaker, LG AI Research trained and deployed its large-scale EXAONEPath model within eight months, using 285 million data points and more than 35,000 high-resolution tissue sample images. Processing and training AI models with extremely large datasets requires immense storage, high-speed data transfer, and significant compute power. With AWS and NVIDIA GPUs, LG AI Research is accelerating training and inference for its deep learning workloads.
LG AI Research uses Amazon S3 to store and retrieve massive volumes of data that are crucial for research. Amazon FSx for Lustre provides sub-millisecond latencies and delivers hundreds of gigabytes per second of throughput, essential for applications that require rapid access to large datasets. This high-performance file and storage system enables parallel data processing and analysis, significantly reducing the time needed to gain insights.
“The healthcare industry is making rapid progress in its use of AI on AWS to accelerate diagnoses and get patients into treatment faster,” said Dan Sheeran, general manager, Healthcare and Life Sciences at AWS. “Using AWS, LG AI Research can develop and use EXAONEPath at an unprecedented scale, reducing data processing and model training times and improving accuracy. This will allow healthcare providers to improve cancer diagnoses and treatments, reduce wait times, and personalize patient care.”
EXAONEPath is part of LG AI Research’s EXAONE, a 300-billion-parameter multimodal foundation model by LG AI Research, which was also built on Amazon SageMaker and Amazon FSx for Lustre. LG AI Research will continue to update and improve EXAONEPath by training it to detect more types of cancer using additional pathology images.
About Amazon Web Services
Since 2006, Amazon Web Services has been the world’s most comprehensive and broadly adopted cloud. AWS has been continually expanding its services to support virtually any workload, and it now has more than 240 fully featured services for compute, storage, databases, networking, analytics, machine learning and artificial intelligence (AI), Internet of Things (IoT), mobile, security, hybrid, media, and application development, deployment, and management from 108 Availability Zones within 34 geographic regions, with announced plans for 18 more Availability Zones and six more AWS Regions in
About Amazon
Amazon is guided by four principles: customer obsession rather than competitor focus, passion for invention, commitment to operational excellence, and long-term thinking. Amazon strives to be Earth’s Most Customer-Centric Company, Earth’s Best Employer, and Earth’s Safest Place to Work. Customer reviews, 1-Click shopping, personalized recommendations, Prime, Fulfillment by Amazon, AWS, Kindle Direct Publishing, Kindle, Career Choice, Fire tablets, Fire TV, Amazon Echo, Alexa, Just Walk Out technology, Amazon Studios, and The Climate Pledge are some of the things pioneered by Amazon. For more information, visit amazon.com/about and follow @AmazonNews.
About LG AI Research
Launched in December 2020 as the artificial intelligence (AI) research think tank of
View source version on businesswire.com: https://www.businesswire.com/news/home/20241203624472/en/
Amazon.com, Inc.
Media Hotline
Amazon-pr@amazon.com
www.amazon.com/pr
Source: Amazon.com, Inc.
FAQ
What is the accuracy rate of LG's EXAONEPath cancer diagnosis model on AWS (AMZN)?
How much faster is genetic testing with LG's new AI model on AWS (AMZN)?
What cost savings does LG AI Research achieve using AWS (AMZN) for their cancer diagnosis model?
