QOCA aim, the AI aided medical imaging and automatic inference platform is now serving a crucial part in our Smart Hospital Solution, since within this platform physicians would be able to efficiently co-work to conduct data processing, data training and furthermore to generate clinical meaningful results for diagnosis.

* Data SystemIntegration

. HIS/PACS auto-manual data exchange mechanism

. DICOM auto-manual exchange mechanism

. DICOM standard protocols for image exchange

. RDB auto-manual data exchange mechanism

. Support for importing MAT/ROI/NII format

. Support for exporting NII format





* Data Processing

. Visualization of feature engineering

. Data preprocessing (missing data, one-hot encoding,

filtering)

. Target value chosen

. Keep feature engineering steps

. Descriptive statistics

. Data distribution

. Data preprocessing steps sharing



* Collaborative image annotation

. Image polygon annotation

. Auto adjustment annotation

. Organ and tumor auto segmentation

. Multi-person collaborative annotation

. Label and annotation feedback mechanism

. Label and annotation management

. MAT/ROI/NII format visualization



* High availability

. Multiple data format supported

. AI model quick start wizard

. Highly flexible in Python development mode

. Medical image and RDB data application integration

. GPU/CPU server management



* High stability, reliability and security

. High scalability

. Container migration easily

. Image/RDB data usage and sharing control

. Role-based authorization

. Custom user roles



* AI model training and development

. AI model auto training mode

. Python development mode

. Multiple programming language supported

. Popular framework supported

. Jupyter Notebook supported

. Data loading/splitting/augmentation

. Hyper parameters/AI model version control

. Varied ML/DL algorithms supported