Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

General General

Applications of artificial intelligence and deep learning in molecular imaging and radiotherapy.

In European journal of hybrid imaging

This brief review summarizes the major applications of artificial intelligence (AI), in particular deep learning approaches, in molecular imaging and radiation therapy research. To this end, the applications of artificial intelligence in five generic fields of molecular imaging and radiation therapy, including PET instrumentation design, PET image reconstruction quantification and segmentation, image denoising (low-dose imaging), radiation dosimetry and computer-aided diagnosis, and outcome prediction are discussed. This review sets out to cover briefly the fundamental concepts of AI and deep learning followed by a presentation of seminal achievements and the challenges facing their adoption in clinical setting.

Arabi Hossein, Zaidi Habib


Artificial intelligence, Deep learning, Molecular imaging, Quantitative imaging, Radiation therapy

General General

B-line quantification: comparing learners novice to lung ultrasound assisted by machine artificial intelligence technology to expert review.

In The ultrasound journal

BACKGROUND : The goal of this study was to assess the ability of machine artificial intelligence (AI) to quantitatively assess lung ultrasound (LUS) B-line presence using images obtained by learners novice to LUS in patients with acute heart failure (AHF), compared to expert interpretation.

METHODS : This was a prospective, multicenter observational study conducted at two urban academic institutions. Learners novice to LUS completed a 30-min training session on lung image acquisition which included lecture and hands-on patient scanning. Learners independently acquired images on patients with suspected AHF. Automatic B-line quantification was obtained offline after completion of the study. Machine AI counted the maximum number of B-lines visualized during a clip. The criterion standard for B-line counts was semi-quantitative analysis by a blinded point-of-care LUS expert reviewer. Image quality was blindly determined by an expert reviewer. A second expert reviewer blindly determined B-line counts and image quality. Intraclass correlation was used to determine agreement between machine AI and expert, and expert to expert.

RESULTS : Fifty-one novice learners completed 87 scans on 29 patients. We analyzed data from 611 lung zones. The overall intraclass correlation for agreement between novice learner images post-processed with AI technology and expert review was 0.56 (confidence interval [CI] 0.51-0.62), and 0.82 (CI 0.73-0.91) between experts. Median image quality was 4 (on a 5-point scale), and correlation between experts for quality assessment was 0.65 (CI 0.48-0.82).

CONCLUSION : After a short training session, novice learners were able to obtain high-quality images. When the AI deep learning algorithm was applied to those images, it quantified B-lines with moderate-to-fair correlation as compared to semi-quantitative analysis by expert review. This data shows promise, but further development is needed before widespread clinical use.

Russell Frances M, Ehrman Robert R, Barton Allen, Sarmiento Elisa, Ottenhoff Jakob E, Nti Benjamin K


Acute heart failure, Artificial intelligence, Lung ultrasound, Novice learner, Point-of-care ultrasound

Radiology Radiology

Deciphering musculoskeletal artificial intelligence for clinical applications: how do I get started?

In Skeletal radiology

Artificial intelligence (AI) represents a broad category of algorithms for which deep learning is currently the most impactful. When electing to begin the process of building an adequate fundamental knowledge base allowing them to decipher machine learning research and algorithms, clinical musculoskeletal radiologists currently have few options to turn to. In this article, we provide an introduction to the vital terminology to understand, how to make sense of data splits and regularization, an introduction to the statistical analyses used in AI research, a primer on what deep learning can or cannot do, and a brief overview of clinical integration methods. Our goal is to improve the readers' understanding of this field.

Mutasa Simukayi, Yi Paul H


Artificial intelligence, Deep learning, Introduction, Machine learning, Musculoskeletal radiology

General General

[Discussion on current state and research strategies of inter-disciplines of acupuncture-moxibustion and artificial intelligence].

In Zhen ci yan jiu = Acupuncture research

At present, the breakthrough of the key techniques of artificial intelligence (AI), including image recognition, deep learning, neural network, Robot technique, etc., greatly promote the development of discipline crossing and medicine. In the present paper, we make an in-depth discussion about the future application of inter-discipline techniques of acupuncture-moxibustion and AI. We think that some of the current instruments have been part of the new acupuncture-moxibustion devices and may have the potential to intersect with the AI discipline. Relying on these existing devices and those of meridian detection, we can obtain relatively objective data, and further conduct meridian-syndrome differentiation and big data collection to possibly realize remote medical treatment. In addition, we may also develop an AI system for studying the underlying mechanisms of acupuncture and moxibustion therapies. Nevertheless, there still exist a lot of problems and challenges in clinical, teaching and scientific researches. It is recommended that the discipline of acupuncture-moxibustion should be intersected with the AI subject and formulate appropriate development strategies, promoting a faster and better development of acupuncturology.

Wu Dong, Sun Han-Xu, Rong Pei-Jing, Dai Ru-Jun, Lian Hai-Hong, Sun Jing-Yi


Acupuncture and moxibustion, Acupuncture device, Artificial intelligence, Development stra-tegy, Inter-discipline

General General

[Review on characteristics of acupuncture-activated network regulatory effect based on brain connectomics].

In Zhen ci yan jiu = Acupuncture research

The network analysis method based on brain connectomics is an important entry point to explore the working mechanism of brain and is also the current trend of researches on acupuncture stimulation-induced changes of neuroimages. We, in the present review, summarized the common network analysis methods for exploring the underlying mechanisms of brain network-mediated regulatory effects of acupuncture interventions. Moreover, combining the current research development and our team's previous research findings, we extracted the characteristics of targeting, conditionity and dynamic of regulatory effects of acupuncture-activated brain network, and put forward our prospects about the future research from the aspects of new scanning modes (for instance, multimodal data acquisition of magnetic resonance, electroencephalogram, near infrared spectrum, etc.), network analysis (such as Granger causality analysis, complex network measures, whole-brain connectivity dynamics tracking, high-order resting-state functional connectivity analysis, etc.) and experimental research paradigms (for example, introduce of transcranial magnetic stimulation induced transient changes of brain functional activity, machine learning approach, etc.).

Xie Kun-Nan, Yin Tao, Ma Pei-Hong, Chen Li, Sun Rui-Rui, Zeng Fang


Acupuncture, Brain network, Connectomics, Regulatory effect, Review

General General

Patients with amnestic MCI Fail to Adapt Executive Control When Repeatedly Tested with Semantic Verbal Fluency Tasks.

In Journal of the International Neuropsychological Society : JINS

OBJECTIVE : Semantic verbal fluency (SVF) tasks require individuals to name items from a specified category within a fixed time. An impaired SVF performance is well documented in patients with amnestic Mild Cognitive Impairment (aMCI). The two leading theoretical views suggest either loss of semantic knowledge or impaired executive control to be responsible.

METHOD : We assessed SVF 3 times on 2 consecutive days in 29 healthy controls (HC) and 29 patients with aMCI with the aim to answer the question which of the two views holds true.

RESULTS : When doing the task for the first time, patients with aMCI produced fewer and more common words with a shorter mean response latency. When tested repeatedly, only healthy volunteers increased performance. Likewise, only the performance of HC indicated two distinct retrieval processes: a prompt retrieval of readily available items at the beginning of the task and an active search through semantic space towards the end. With repeated assessment, the pool of readily available items became larger in HC, but not patients with aMCI.

CONCLUSION : The production of fewer and more common words in aMCI points to a smaller search set and supports the loss of semantic knowledge view. The failure to improve performance as well as the lack of distinct retrieval processes point to an additional impairment in executive control. Our data did not clearly favour one theoretical view over the other, but rather indicates that the impairment of patients with aMCI in SVF is due to a combination of both.

Tröger Johannes, Lindsay Hali, Mina Mario, Linz Nicklas, Klöppel Stefan, Kray Jutta, Peter Jessica


Amnestic MCI, Executive control, Practice effects, Semantic loss, Semantic verbal fluency, Temporal analysis