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

Radiology Radiology

Implementation of artificial intelligence (AI) applications in radiology: hindering and facilitating factors.

In European radiology ; h5-index 62.0

OBJECTIVE : The objective was to identify barriers and facilitators to the implementation of artificial intelligence (AI) applications in clinical radiology in The Netherlands.

MATERIALS AND METHODS : Using an embedded multiple case study, an exploratory, qualitative research design was followed. Data collection consisted of 24 semi-structured interviews from seven Dutch hospitals. The analysis of barriers and facilitators was guided by the recently published Non-adoption, Abandonment, Scale-up, Spread, and Sustainability (NASSS) framework for new medical technologies in healthcare organizations.

RESULTS : Among the most important facilitating factors for implementation were the following: (i) pressure for cost containment in the Dutch healthcare system, (ii) high expectations of AI's potential added value, (iii) presence of hospital-wide innovation strategies, and (iv) presence of a "local champion." Among the most prominent hindering factors were the following: (i) inconsistent technical performance of AI applications, (ii) unstructured implementation processes, (iii) uncertain added value for clinical practice of AI applications, and (iv) large variance in acceptance and trust of direct (the radiologists) and indirect (the referring clinicians) adopters.

CONCLUSION : In order for AI applications to contribute to the improvement of the quality and efficiency of clinical radiology, implementation processes need to be carried out in a structured manner, thereby providing evidence on the clinical added value of AI applications.

KEY POINTS : • Successful implementation of AI in radiology requires collaboration between radiologists and referring clinicians. • Implementation of AI in radiology is facilitated by the presence of a local champion. • Evidence on the clinical added value of AI in radiology is needed for successful implementation.

Strohm Lea, Hehakaya Charisma, Ranschaert Erik R, Boon Wouter P C, Moors Ellen H M


Artificial intelligence, Computer systems, Computer-assisted, Diagnosis, Information systems, Radiology

Public Health Public Health

Hybrid Stem Cell States: Insights Into the Relationship Between Mammary Development and Breast Cancer Using Single-Cell Transcriptomics.

In Frontiers in cell and developmental biology

Similarities between stem cells and cancer cells have implicated mammary stem cells in breast carcinogenesis. Recent evidence suggests that normal breast stem cells exist in multiple phenotypic states: epithelial, mesenchymal, and hybrid epithelial/mesenchymal (E/M). Hybrid E/M cells in particular have been implicated in breast cancer metastasis and poor prognosis. Mounting evidence also suggests that stem cell phenotypes change throughout the life course, for example, through embryonic development and pregnancy. The goal of this study was to use single cell RNA-sequencing to quantify cell state distributions of the normal mammary (NM) gland throughout developmental stages and when perturbed into a stem-like state in vitro using conditional reprogramming (CR). Using machine learning based dataset alignment, we integrate multiple mammary gland single cell RNA-seq datasets from human and mouse, along with bulk RNA-seq data from breast tumors in the Cancer Genome Atlas (TCGA), to interrogate hybrid stem cell states in the normal mammary gland and cancer. CR of human mammary cells induces an expanded stem cell state, characterized by increased expression of embryonic stem cell associated genes. Alignment to a mouse single-cell transcriptome atlas spanning mammary gland development from in utero to adulthood revealed that NM cells align to adult mouse cells and CR cells align across the pseudotime trajectory with a stem-like population aligning to the embryonic mouse cells. Three hybrid populations emerge after CR that are rare in NM: KRT18+/KRT14+ (hybrid luminal/basal), EPCAM+/VIM+ (hybrid E/M), and a quadruple positive population, expressing all four markers. Pseudotime analysis and alignment to the mouse developmental trajectory revealed that E/M hybrids are the most developmentally immature. Analyses of single cell mouse mammary RNA-seq throughout pregnancy show that during gestation, there is an enrichment of hybrid E/M cells, suggesting that these cells play an important role in mammary morphogenesis during lactation. Finally, pseudotime analysis and alignment of TCGA breast cancer expression data revealed that breast cancer subtypes express distinct developmental signatures, with basal tumors representing the most "developmentally immature" phenotype. These results highlight phenotypic plasticity of normal mammary stem cells and provide insight into the relationship between hybrid cell populations, stemness, and cancer.

Thong Tasha, Wang Yutong, Brooks Michael D, Lee Christopher T, Scott Clayton, Balzano Laura, Wicha Max S, Colacino Justin A


breast cancer, epithelial, hybrid, mesenchymal, pregnancy, single-cell RNA sequencing, stem cells

General General

Lower Limb Kinematics Trajectory Prediction Using Long Short-Term Memory Neural Networks.

In Frontiers in bioengineering and biotechnology

This study determined whether the kinematics of lower limb trajectories during walking could be extrapolated using long short-term memory (LSTM) neural networks. It was hypothesised that LSTM auto encoders could reliably forecast multiple time-step trajectories of the lower limb kinematics, specifically linear acceleration (LA) and angular velocity (AV). Using 3D motion capture, lower limb position-time coordinates were sampled (100 Hz) from six male participants (age 22 ± 2 years, height 1.77 ± 0.02 m, body mass 82 ± 4 kg) who walked for 10 min at 5 km/h on a 0% gradient motor-driven treadmill. These data were fed into an LSTM model with a sliding window of four kinematic variables with 25 samples or time steps: LA and AV for thigh and shank. The LSTM was tested to forecast five samples (i.e., time steps) of the four kinematic input variables. To attain generalisation, the model was trained on a dataset of 2,665 strides from five participants and evaluated on a test set of 1 stride from a sixth participant. The LSTM model learned the lower limb kinematic trajectories using the training samples and tested for generalisation across participants. The forecasting horizon suggested higher model reliability in predicting earlier future trajectories. The mean absolute error (MAE) was evaluated on each variable across the single tested stride, and for the five-sample forecast, it obtained 0.047 m/s2 thigh LA, 0.047 m/s2 shank LA, 0.028 deg/s thigh AV and 0.024 deg/s shank AV. All predicted trajectories were highly correlated with the measured trajectories, with correlation coefficients greater than 0.98. The motion prediction model may have a wide range of applications, such as mitigating the risk of falls or balance loss and improving the human-machine interface for wearable assistive devices.

Zaroug Abdelrahman, Lai Daniel T H, Mudie Kurt, Begg Rezaul


LSTM, forecasting, gait, machine learning, neural networks, walking

General General

Extracellular Vesicles in Renal Cell Carcinoma: Multifaceted Roles and Potential Applications Identified by Experimental and Computational Methods.

In Frontiers in oncology

Renal cell carcinoma (RCC) is the most common type of kidney cancer. Increasingly evidences indicate that extracellular vesicles (EVs) orchestrate multiple processes in tumorigenesis, metastasis, immune evasion, and drug response of RCC. EVs are lipid membrane-bound vesicles in nanometer size and secreted by almost all cell types into the extracellular milieu. A myriad of bioactive molecules such as RNA, DNA, protein, and lipid are able to be delivered via EVs for the intercellular communication. Hence, the abundant content of EVs is appealing reservoir for biomarker identification through computational analysis and experimental validation. EVs with excellent biocompatibility and biodistribution are natural platforms that can be engineered to offer achievable drug delivery strategies for RCC therapies. Moreover, the multifaceted roles of EVs in RCC progression also provide substantial targets and facilitate EVs-based drug discovery, which will be accelerated by using artificial intelligence approaches. In this review, we summarized the vital roles of EVs in occurrence, metastasis, immune evasion, and drug resistance of RCC. Furthermore, we also recapitulated and prospected the EVs-based potential applications in RCC, including biomarker identification, drug vehicle development as well as drug target discovery.

Qin Zhiyuan, Xu Qingwen, Hu Haihong, Yu Lushan, Zeng Su


artificial intelligence, biomarkers, drug targets, drug vehicles, exosomes, extracellular vesicles, machine learning, renal cell carcinoma

General General

Digital Innovations for Global Mental Health: Opportunities for Data Science, Task Sharing, and Early Intervention.

In Current treatment options in psychiatry

Purpose : Globally, individuals living with mental disorders are more likely to have access to a mobile phone than mental health care. In this commentary, we highlight opportunities for expanding access to and use of digital technologies to advance research and intervention in mental health, with emphasis on the potential impact in lower resource settings.

Recent findings : Drawing from empirical evidence, largely from higher income settings, we considered three emerging areas where digital technology will potentially play a prominent role: supporting methods in data science to further our understanding of mental health and inform interventions, task sharing for building workforce capacity by training and supervising non-specialist health workers, and facilitating new opportunities for early intervention for young people in lower resource settings. Challenges were identified related to inequities in access, threats of bias in big data analyses, risks to users, and need for user involvement to support engagement and sustained use of digital interventions.

Summary : For digital technology to achieve its potential to transform the ways we detect, treat, and prevent mental disorders, there is a clear need for continued research involving multiple stakeholders, and rigorous studies showing that these technologies can successfully drive measurable improvements in mental health outcomes.

Naslund John A, Gonsalves Pattie P, Gruebner Oliver, Pendse Sachin R, Smith Stephanie L, Sharma Amit, Raviola Giuseppe


Artificial intelligence, Big data, Digital technology, Early intervention, Global mental health, Task sharing

General General

Circular RNA in Schizophrenia and Depression.

In Frontiers in psychiatry

Schizophrenia (SZ) and depression (DEP) are two common major psychiatric disorders that are associated with high risk of suicide. These disorders affect not only physical and mental health, but they also affect the social function of the individual. However, diagnoses of SZ and DEP are mainly based on symptomatic changes and the clinical experience of psychiatrists. These rather subjective measures can induce misdiagnoses and missed diagnoses. Therefore, it is necessary to further explore objective indexes for improving the early diagnoses and prognoses of SZ and DEP. Current research indicates that non-coding RNA (ncRNA) may play a role in the occurrence and development of SZ and DEP. Circular RNA (circRNA), as an important component of ncRNA, is associated with many biological functions, especially post-transcriptional regulation. Since circRNA is easily detected in peripheral blood and has a high degree of spatiotemporal tissue specificity and stability, these attributes provide us with a new idea to further explore the potential value for the diagnosis and treatment of SZ and DEP. Here, we summarize the classification, characteristics, and biological functions of circRNA and the most significant results of experimental studies, aiming to highlight the involvement of circRNA in SZ and DEP.

Li Zexuan, Liu Sha, Li Xinrong, Zhao Wentao, Li Jing, Xu Yong


biological function, circular RNA (circRNA), depression (DEP), epigenetic characteristics, expression, schizophrenia (SZ)