Deep learning techniques have been shown to successfully address many medical image technical problems, e.g., segmentation, lesion detection, image registration, etc., better than ever. Accordingly, the development of artificial intelligence (AI) applications in medicine is continuously booming in these years for the attempt to improve various image-related medical procedures and workflows. The applications aim to improve the workup of image diagnosis, optimize the imaging workflow, offload the tedious medical tasks, and provide richer and more holistic cues for treatment planning. Specific prominent applications may range from lesion detections like pulmonary nodules and breast lesions, cancer risk prediction, triaging, abnormality quantification and measurement, disease staging, quality control, imaging improvement, surgical/therapy planning, etc. Thus far, many commercial AI applications have been incorporated into clinical practice in this decade. Meanwhile, many promising clinical validation studies were also published in recent years with a vast amount of evaluation data across multiple institutes and countries. With the further breakthrough of ChatGPT, the trend of AI for medicine seems to be unstoppable and will persist in advancing. However, there are still plenty of unsolved or partially solved medical problems with which AI techniques may help.
This Research Topic aims to explore AI/machine learning applications for the improvement of medical workflow, procedure, diagnostic workup, etc., that relate to radiological images. The potential applications may include new imaging methods for better efficiency or image quality, workflow optimization for quality control and image reading, workup improvement for image diagnosis, cancer risk prediction, supportive visualization and analytic tools for treatment procedures, like planning, follow-up, and so on. The content of an AI/machine learning application can be technological development, clinical evaluation, or the exploitation of new directions that could change the current paradigm of medical image-related processes. In particular, AI applications equipped with the cutting-edge generative pre-trained transformer (GPT) as well as the exploratory image-related scenes incorporated with the question and answer (Q&A) scheme, empowered by ChatGPT and other advanced language models, are also highly welcome.
We call for Original Research, Methods, Systematic Review, Perspective, Clinical Trials, Case Reports, and Brief Research Reports about AI/machine applications for medical image-related workflows. The potential themes are shown below, but not limited to:
- Computer-aided detection of abnormalities in medical images
- Computer-aided diagnosis
- Risk prediction for cancers
- Computer-aided surgical or treatment planning
- Medical workflow optimization
- Computer-aided image quality control
- New medical imaging methods with AI techniques
- Triaging for critical findings
- Prognosis prediction with image cues
- Clinical evaluation for AI applications
Topic Editor Jie-Zhi Cheng is is employed by Shanghai United Imaging Intelligence Co., Ltd. Other Topic Editors declare no conflicts of interest with the Research Topic.
Deep learning techniques have been shown to successfully address many medical image technical problems, e.g., segmentation, lesion detection, image registration, etc., better than ever. Accordingly, the development of artificial intelligence (AI) applications in medicine is continuously booming in these years for the attempt to improve various image-related medical procedures and workflows. The applications aim to improve the workup of image diagnosis, optimize the imaging workflow, offload the tedious medical tasks, and provide richer and more holistic cues for treatment planning. Specific prominent applications may range from lesion detections like pulmonary nodules and breast lesions, cancer risk prediction, triaging, abnormality quantification and measurement, disease staging, quality control, imaging improvement, surgical/therapy planning, etc. Thus far, many commercial AI applications have been incorporated into clinical practice in this decade. Meanwhile, many promising clinical validation studies were also published in recent years with a vast amount of evaluation data across multiple institutes and countries. With the further breakthrough of ChatGPT, the trend of AI for medicine seems to be unstoppable and will persist in advancing. However, there are still plenty of unsolved or partially solved medical problems with which AI techniques may help.
This Research Topic aims to explore AI/machine learning applications for the improvement of medical workflow, procedure, diagnostic workup, etc., that relate to radiological images. The potential applications may include new imaging methods for better efficiency or image quality, workflow optimization for quality control and image reading, workup improvement for image diagnosis, cancer risk prediction, supportive visualization and analytic tools for treatment procedures, like planning, follow-up, and so on. The content of an AI/machine learning application can be technological development, clinical evaluation, or the exploitation of new directions that could change the current paradigm of medical image-related processes. In particular, AI applications equipped with the cutting-edge generative pre-trained transformer (GPT) as well as the exploratory image-related scenes incorporated with the question and answer (Q&A) scheme, empowered by ChatGPT and other advanced language models, are also highly welcome.
We call for Original Research, Methods, Systematic Review, Perspective, Clinical Trials, Case Reports, and Brief Research Reports about AI/machine applications for medical image-related workflows. The potential themes are shown below, but not limited to:
- Computer-aided detection of abnormalities in medical images
- Computer-aided diagnosis
- Risk prediction for cancers
- Computer-aided surgical or treatment planning
- Medical workflow optimization
- Computer-aided image quality control
- New medical imaging methods with AI techniques
- Triaging for critical findings
- Prognosis prediction with image cues
- Clinical evaluation for AI applications
Topic Editor Jie-Zhi Cheng is is employed by Shanghai United Imaging Intelligence Co., Ltd. Other Topic Editors declare no conflicts of interest with the Research Topic.