Related Experiment Video
Updated: Jan 25, 2026
Identification and Protection of the Recurrent Laryngeal Nerve during Transoral Robotic Thyroidectomy
Published on: October 24, 2025
Precise diagnosis of intracranial hemorrhage and subtypes using a three-dimensional joint convolutional and recurrent
Hai Ye1, Feng Gao2, Youbing Yin2
1Department of Radiology, Shenzhen Second People's Hospital, Shenzhen Second Hospital Clinical Medicine College of Anhui Medical University, Shenzhen, China.
Insights
A novel 3D deep learning model accurately detects intracranial hemorrhage (ICH) and its subtypes on CT scans. This artificial intelligence tool demonstrates superior performance compared to junior radiologists, aiding clinical diagnosis.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Radiology
Background:
- Intracranial hemorrhage (ICH) is a critical condition requiring rapid diagnosis.
- Accurate identification of ICH subtypes is essential for effective treatment.
Purpose of the Study:
- To evaluate a novel 3D joint convolutional and recurrent neural network (CNN-RNN) for detecting ICH and its subtypes.
- To assess the performance of the deep learning model against human interpretations.
Main Methods:
- A retrospective study included 2836 subjects with non-contrast head CT scans.
- A 3D joint CNN-RNN framework was developed for ICH detection and subtype classification.
- The model's predictions were compared with radiologist interpretations.
Main Results:
- The algorithm processed 3D CT scans in under 30 seconds.
- Excellent performance (≥0.98 metrics) for binary ICH detection.
- Achieved >0.8 AUC for five-subtype classification, outperforming junior trainees.
Conclusions:
- The 3D CNN-RNN model accurately and rapidly detects ICH and its subtypes.
- The deep learning framework shows potential to assist radiologists in clinical workflows.
- The model's performance suggests it can help reduce initial misinterpretations in ICH diagnosis.
Objectives:
To evaluate the performance of a novel three-dimensional (3D) joint convolutional and recurrent neural network (CNN-RNN) for the detection of intracranial hemorrhage (ICH) and its five subtypes (cerebral parenchymal, intraventricular, subdural, epidural, and subarachnoid) in non-contrast head CT.
Methods:
A total of 2836 subjects (ICH/normal, 1836/1000) from three institutions were included in this ethically approved retrospective study, with a total of 76,621 slices from non-contrast head CT scans. ICH and its five subtypes were annotated by three independent experienced radiologists, with majority voting as reference standard for both the subject level and the slice level. Ninety percent of data was used for training and validation, and the rest 10% for final evaluation. A joint CNN-RNN classification framework was proposed, with the flexibility to train when subject-level or slice-level labels are available. The predictions were compared with the interpretations from three junior radiology trainees and an additional senior radiologist.
Results:
It took our algorithm less than 30 s on average to process a 3D CT scan. For the two-type classification task (predicting bleeding or not), our algorithm achieved excellent values (≥ 0.98) across all reporting metrics on the subject level. For the five-type classification task (predicting five subtypes), our algorithm achieved > 0.8 AUC across all subtypes. The performance of our algorithm was generally superior to the average performance of the junior radiology trainees for both two-type and five-type classification tasks.
Conclusions:
The proposed method was able to accurately detect ICH and its subtypes with fast speed, suggesting its potential for assisting radiologists and physicians in their clinical diagnosis workflow.
Key Points:
• A 3D joint CNN-RNN deep learning framework was developed for ICH detection and subtype classification, which has the flexibility to train with either subject-level labels or slice-level labels. • This deep learning framework is fast and accurate at detecting ICH and its subtypes. • The performance of the automated algorithm was superior to the average performance of three junior radiology trainees in this work, suggesting its potential to reduce initial misinterpretations.
Related Concept Videos
01:17Convolution Properties II
The width property indicates that if the durations of input signals are T1 and T2, then the width of the output response equals the sum of both durations, irrespective of the shapes of the two functions. For instance, convolving two rectangular pulses with durations of 2 seconds and 1 second results in a function with a width of 3 seconds.
The area property asserts that the area under the...
02:18Network Covalent Solids
To break or to melt a covalent network solid, covalent bonds must be broken. Because covalent bonds are relatively strong, covalent network solids are typically...
01:20Convolution Properties I
The commutative property reveals that the input and the impulse response of an LTI (Linear Time-Invariant) system can be interchanged without affecting the output:
01:16Structural Joints: Synovial Joints
01:03Structural Joints: Fibrous Joints
Suture
All the bones of the skull, except for the mandible, are joined to each other by a fibrous joint called a suture. The fibrous connective tissue found at a suture strongly unites the adjacent skull bones and thus helps to protect the brain and form the face. In...
01:17Structural Joints: Cartilaginous Joints
There are two types of cartilaginous joints:
Synchondrosis
A synchondrosis ("joined by cartilage") is a cartilaginous joint where bones are connected by hyaline cartilage. Synchondrosis may be temporary...