Search

The utility of Artificial Intelligence (AI) in healthcare strongly depends upon the quality of the data used to build models, and the confidence in the predictions they generate. Access to sufficient amounts of high-quality data to build accurate and reliable models remains problematic owing to substantive legal and ethical constraints in making clinically relevant research data available offsite. New technologies such as distributed learning offer a pathway forward, but unfortunately tend to suffer from a lack of transparency, which undermines trust in what data are used for the analysis. To address such issues, we hypothesized that, a novel distributed learning that combines sequential distributed learning with a blockchain-based platform, namely Chained Distributed Machine learning C-DistriM, would be feasible and would give a similar result as a standard centralized approach. C-DistriM enables health centers to dynamically participate in training distributed learning models. We demonstrate C-DistriM using the NSCLC-Radiomics open data to predict two-year lung-cancer survival. A comparison of the performance of this distributed solution, evaluated in six different scenarios, and the centralized approach, showed no statistically significant difference (AUCs between central and distributed models), all DeLong tests yielded p -val >0.05. This methodology removes the need to blindly trust the computation in one specific server on a distributed learning network. This fusion of blockchain and distributed learning serves as a proof-of-concept to increase transparency, trust, and ultimately accelerate the adoption of AI in multicentric studies. We conclude that our blockchain-based model for sequential training on distributed datasets is a feasible approach, provides equivalent performance to the centralized approach.

During the COVID-19 pandemic induced by the SARS-CoV-2, numerous chest scans were carried out in order to establish the diagnosis, quantify the extension of lesions but also identify the occurrence of potential pulmonary embolisms. In this perspective, the performed chest scans provided a varied database for a retrospective analysis of non-COVID-19 chest pathologies discovered de novo. The fortuitous discovery of de novo non-COVID-19 lesions was generally not detected by the automated systems for COVID-19 pneumonia developed in parallel during the pandemic and was thus identified on chest CT by the radiologist. The objective is to use the study of the occurrence of non-COVID-19-related chest abnormalities (known and unknown) in a large cohort of patients having suffered from confirmed COVID-19 infection and statistically correlate the clinical data and the occurrence of these abnormalities in order to assess the potential of increased early detection of lesions/alterations. This study was performed on a group of 362 COVID-19-positive patients who were prescribed a CT scan in order to diagnose and predict COVID-19-associated lung disease. Statistical analysis using mean, standard deviation (SD) or median and interquartile range (IQR), logistic regression models and linear regression models were used for data analysis. Results were considered significant at the 5% critical level (p < 0.05). These de novo non-COVID-19 thoracic lesions detected on chest CT showed a significant prevalence in cardiovascular pathologies, with calcifying atheromatous anomalies approaching nearly 35.4% in patients over 65 years of age. The detection of non-COVID-19 pathologies was mostly already known, except for suspicious nodule, thyroid goiter and the ascending thoracic aortic aneurysm. The presence of vertebral compression or signs of pulmonary fibrosis has shown a significant impact on inpatient length of stay. The characteristics of the patients in this sample, both from a demographic and a tomodensitometric point of view on non-COVID-19 pathologies, influenced the length of hospital stay as well as the risk of intra-hospital death. This retrospective study showed that the potential importance of the detection of these non-COVID-19 lesions by the radiologist was essential in the management and the intra-hospital course of the patients.

The coronavirus disease 2019 (COVID-19) outbreak has reached pandemic status. Drastic measures of social distancing are enforced in society and healthcare systems are being pushed to and beyond their limits. To help in the fight against this threat on human health, a fully automated AI framework was developed to extract radiomics features from volumetric chest computed tomography (CT) exams. The detection model was developed on a dataset of 1381 patients (181 COVID-19 patients plus 1200 non COVID control patients). A second, independent dataset of 197 RT-PCR confirmed COVID-19 patients and 500 control patients was used to assess the performance of the model. Diagnostic performance was assessed by the area under the receiver operating characteristic curve (AUC). The model had an AUC of 0.882 (95% CI: 0.851–0.913) in the independent test dataset (641 patients). The optimal decision threshold, considering the cost of false negatives twice as high as the cost of false positives, resulted in an accuracy of 85.18%, a sensitivity of 69.52%, a specificity of 91.63%, a negative predictive value (NPV) of 94.46% and a positive predictive value (PPV) of 59.44%. Benchmarked against RT-PCR confirmed cases of COVID-19, our AI framework can accurately differentiate COVID-19 from routine clinical conditions in a fully automated fashion. Thus, providing rapid accurate diagnosis in patients suspected of COVID-19 infection, facilitating the timely implementation of isolation procedures and early intervention.

The information that connects genotypes and phenotypes is essentially embedded in research articles written in natural language. To facilitate access to this knowledge, we constructed a framework for the curation of the scientific literature studying the molecular mechanisms that control leaf growth and development in Arabidopsis thaliana (Arabidopsis). Standard structured statements, called relations, were designed to capture diverse data types, including phenotypes and gene expression linked to genotype description, growth conditions, genetic and molecular interactions, and details about molecular entities. Relations were then annotated from the literature, defining the relevant terms according to standard biomedical ontologies. This curation process was supported by a dedicated graphical user interface, called Leaf Knowtator. A total of 283 primary research articles were curated by a community of annotators, yielding 9947 relations monitored for consistency and over 12,500 references to Arabidopsis genes. This information was converted into a relational database (KnownLeaf) and merged with other public Arabidopsis resources relative to transcriptional networks, protein–protein interaction, gene co-expression, and additional molecular annotations. Within KnownLeaf, leaf phenotype data can be searched together with molecular data originating either from this curation initiative or from external public resources. Finally, we built a network (LeafNet) with a portion of the KnownLeaf database content to graphically represent the leaf phenotype relations in a molecular context, offering an intuitive starting point for knowledge mining. Literature curation efforts such as ours provide high quality structured information accessible to computational analysis, and thereby to a wide range of applications. DATA: The presented work was performed in the framework of the AGRON-OMICS project (Arabidopsis GRO wth Network integrating OMICS technologies) supported by European Commission 6th Framework Programme project (Grant number LSHG-CT-2006-037704). This is a data integration and data sharing portal collecting all the all the major results from the consortium. All data presented in our paper is available here. https://agronomics.ethz.ch/.

Plants adapt to the prevailing photoperiod by adjusting growth and flowering to the availability of energy. To understand the molecular changes involved in adaptation to a long-day condition we comprehensively profiled leaf six at the end of the day and the end of the night at four developmental stages on Arabidopsis thaliana plants grown in a 16 h photoperiod, and compared the profiles to those from leaf 6 of plants grown in a 8 h photoperiod. When Arabidopsis is grown in a long-day photoperiod individual leaf growth is accelerated but whole plant leaf area is decreased because total number of rosette leaves is restricted by the rapid transition to flowering. Carbohydrate measurements in long- and short-day photoperiods revealed that a long photoperiod decreases the extent of diurnal turnover of carbon reserves at all leaf stages. At the transcript level we found that the long-day condition has significantly reduced diurnal transcript level changes than in short-day condition, and that some transcripts shift their diurnal expression pattern. Functional categorisation of the transcripts with significantly different levels in short and long day conditions revealed photoperiod-dependent differences in RNA processing and light and hormone signalling, increased abundance of transcripts for biotic stress response and flavonoid metabolism in long photoperiods, and for photosynthesis and sugar transport in short photoperiods. Furthermore, we found transcript level changes consistent with an early release of flowering repression in the long-day condition. Differences in protein levels between long and short photoperiods mainly reflect an adjustment to the faster growth in long photoperiods. In summary, the observed differences in the molecular profiles of leaf six grown in long- and short-day photoperiods reveal changes in the regulation of metabolism that allow plants to adjust their metabolism to the available light. The data also suggest that energy management is in the two photoperiods fundamentally different as a consequence of photoperiod-dependent energy constraints.

The circadian clock regulates physiological processes central to growth and survival. To date, most plant circadian clock studies have relied on diurnal transcriptome changes to elucidate molecular connections between the circadian clock and observable phenotypes in wild-type plants. Here, we have integrated RNA-sequencing and protein mass spectrometry data to comparatively analyse the lhycca1, prr7prr9, gi and toc1 circadian clock mutant rosette at the end of day and end of night. Each mutant affects specific sets of genes and proteins, suggesting that the circadian clock regulation is modular. Furthermore, each circadian clock mutant maintains its own dynamically fluctuating transcriptome and proteome profile specific to subcellular compartments. Most of the measured protein levels do not correlate with changes in their corresponding transcripts. Transcripts and proteins that have coordinated changes in abundance are enriched for carbohydrate- and cold-responsive genes. Transcriptome changes in all four circadian clock mutants also affect genes encoding starch degradation enzymes, transcription factors and protein kinases. The comprehensive transcriptome and proteome datasets demonstrate that future system-driven research of the circadian clock requires multi-level experimental approaches. Our work also shows that further work is needed to elucidate the roles of post-translational modifications and protein degradation in the regulation of clock-related processes.

'Water grabbing' and 'land grabbing' have been referred to as a new colonialism, dispossessing small farmers and indigenous people of land and water for the sake of investors. The current 'grabbing' is driven by perceived scarcity of food and sustainable energy, and is enabled by global financial instruments and commodity speculation. In this paper, we argue that while in many ways different, the 'new colonialism' of land/water grabbing may be better understood through analysis of old colonialism. We use actor network and place modernisation theories to analyse the history and practice of Zionist land/water grabbing in Israel/Palestine as an ongoing remnant of old colonialism. While there are clearly unique aspects to this case, there are similarities in processes, such as the narrative of modernising 'barren', 'infertile', and 'undeveloped' land. The ongoing power imbalance in water management and access, the disproportionate burden on Palestinians of growing water scarcity, and the inability of technical fixes to address the problems of relative deprivation may be seen as cautionary tales for current 'water grabbing'.