Articles

Chitalia, Miliotis et al. evaluate radiomic predictors of outcome in patients with breast cancer treated with neoadjuvant chemotherapy. The authors identify radiomic phenotypes related to changes in tumor heterogeneity that improve progression-free survival prediction when added to clinicopathological and molecular factors.

Tangirala et al. evaluate changes in the associations between various exposures and COVID-19 positivity and hospitalization across two non-consecutive waves of the pandemic in participants of the UK Biobank. They find that the strength of the association between certain risk factors, such as age and household-related factors, change over time.

Kiyasseh et al. compare the quality of feedback provided to surgeons by artificial intelligence (AI) to that provided by human experts. Teaching an AI system to explicitly follow human explanations improves the reliability and reduces the bias of AI-based feedback.

Steyaert, Qiu et al. develop a deep learning framework for multimodal data fusion for adult and pediatric brain tumors. Multimodal data models combining histopathology imaging and gene expression data outperform single data models in predicting prognosis.

Mintz et al. develop machine learning models to predict the probability of ciprofloxacin resistance in hospitalized patients. Resistance of previous infections, prior location of patients, sex and recent resistance frequencies in the hospital impact the probability of ciprofloxacin resistance in each patient.

Fukuhara et al. evaluate different methods to express IL-12 as an immunostimulatory payload of oncolytic HSV-1 using G47∆ as the backbone. They find that expressing the two IL-12 subunits as a fusion peptide leads to superior efficacy of the oncolytic virus in tumor models, compared with a virus in which the subunits have been co-expressed.

Ma et al. combine forecasts for influenza-like disease and COVID-19 to model scenarios where these diseases co-exist. They demonstrate that pooling information about influenza-like disease and COVID-19 improves forecasting models compared to individual models for either disease.

Haggenmüller et al. evaluate the performance of a smartphone app for non-invasive hemoglobin estimation that was developed in the USA in rural India. Performance improved when the app was retrained on the data collected in India.

Thomas et al. test 6 ELISAs detecting IgA and IgG antibodies to whole SARS-CoV-2 spike protein, to its receptor binding domain region and to nucleocapsid protein in saliva. Across 20 household outbreaks, antibody responses are heterogeneous, but a reliable indicator of recent infection.

Griffith et al. report COVID-19 symptom persistence in a national population-based cohort over 50 years old in Canada. These data allow the authors to estimate the background symptom prevalence in the cohort, given not everyone had COVID-19, but also to explore a broader range of pre-pandemic risk factors for post-acute COVID-19 symptom persistence.

Srinath, Xie et al. analyze plasma metabolites present in patients with cerebral cavernous angiomas. Cholic acid and hypoxanthine are found in those with Cavernous Angioma disease whilst arachidonic and linoleic acids are found in Cavernous Angioma patients with symptomatic hemorrhage.

Chang et al. classify people with Temporal Lobe Epilepsy (TLE), Alzheimer’s disease and healthy controls using a convolutional neural network algorithm applied to magnetic resonance imaging (MRI) scans. People with TLE can be distinguished, including those without easily identifiable TLE-associated MRI features.

Hirotsu, Kakizaki et al. assess the presence of pneumonia and pathogenesis during infection with different SARS-CoV-2 variants. The prevalence of pneumonia differs depending on the infectious variant, with the lowest prevalence and mild lung pathogenesis following infection with BA.2.

Nyblom, Johnning et al. develop an optical DNA mapping approach for bacterial strain typing of patient samples. They demonstrate rapid identification of clinically relevant E. coli and K. pneumoniae strains, without the need for cultivation.

Roederer et al. estimate uptake of COVID-19 vaccines amongst migrants, homeless and precariously housed people in two regions of France with cross-sectional survey data. They also report sociodemographic factors and reasons associated with (non-)vaccination.

Wang, Liang, Huang et al. evaluate associations between mitochondrial DNA copy number (mtDNA-CN) and HbA1c and weight loss in patients with type 2 diabetes treated with metformin or acarbose, as part of the MARCH study. They find that patients with higher mtDNA-CN lost more weight on metformin, but did not see the same with acarbose.

Nuzzo, Russo, Errico, D’Amico et al. investigate neuroinflammation in forty-eight pediatric spinal muscular atrophy patients before and after Nusinersen treatment. They find signatures of neuroinflammation that are specifically associated with severe disease and show that Nusinersen therapy has neuro-immunomodulatory effects.

Vaid et al performed a multi-center retrospective cohort study using electrocardiograms from patients with mitral regurgitation to train a deep learning model to detect valvular disease and validated it in externally. They demonstrate the model could potentially enable earlier disease detection and improve overall prognosis.

Klein et al. use mobility data to forecast COVID-19 admissions for five Massachusetts hospitals. Combining aggregated mobile device data about users’ contact patterns, commuting volume, and mobility range with COVID hospitalizations and test-positivity data increases the lead-time of accurate predictions for individual hospitals.

Tang, Liang, Shen, et al. assessed whether the use of a health management app by people with Parkinson’s disease before and after a COVID-19 lockdown in China had an impact on their quality of life (QoL). They show that these social distancing measures reduced the QoL overall, but the reduction was less pronounced if adherence to the app was high.