SLSam Lockhart

Sam Lockhart

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Sam Lockhart has a total of 24 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Prone positioning is associated with increased insulin requirements in mechanically ventilated patients with COVID-19

    by Harry Griffiths, Amy Cardwell, Max Richardson, Meg Barne, Bogdan Petrisor, Ammara Usman, Laura Heales, Julia Calvo Latorre, Vishakha Bansiya, Razeen Mahroof, et al.

    Stress hyperglycaemia is common in critical illness. We have previously observed that increasing severity of respiratory failure in patients with severe COVID-19 is associated with increased insulin demand. Given previously reported direct effects of hypoxia on insulin action, we reasoned that rapid improvements in oxygenation following prone positioning may improve insulin sensitivity and increase risk of hypoglycaemia. A retrospective multi-centre service evaluation comparing blood glucose and insulin administration in patients with COVID-19 pneumonitis receiving prone mechanical ventilation, comparing the 16 h pre-prone and 16 h post-prone time periods. 155 patients were included in this analysis. Oxygenation improved significantly following prone positioning (change in SpO2/FIO2 per hour prone: 3.01 ± 0.14, P < 0.0001). Glycaemic control was similar during the supine and prone study periods, and there were no hypoglycaemic events in the prone study period. Prone positioning was associated with an unexpected modest but significant increase in insulin requirements (mean difference in total insulin dose (IU): 8.32 ± 2.14, P < 0.001) that was robust to several sensitivity analyses, and could not be explained by changes in carbohydrate intake. We did not observe an increased rate of hypoglycaemia during prone ventilation and the adequacy of glycaemic control was comparable during the supine and prone study periods. Unexpectedly, prone ventilation was associated with an increase in insulin requirements despite significant improvement in hypoxaemia. Our findings support the safety of prone ventilation with respect to glycaemic control and identify a novel relationship between ventilation position and insulin requirements in critical illness.

  • Open Access

    Usefulness of Heat Map Explanations for Deep-Learning-Based Electrocardiogram Analysis

    by Andrea M. Storås, Ole Emil Andersen, Sam Lockhart, Roman Thielemann, Filip Gnesin, Vajira Thambawita, Steven A. Hicks, Jørgen K. Kanters, Inga Strümke, Pål Halvorsen, et al.

    Deep neural networks are complex machine learning models that have shown promising results in analyzing high-dimensional data such as those collected from medical examinations. Such models have the potential to provide fast and accurate medical diagnoses. However, the high complexity makes deep neural networks and their predictions difficult to understand. Providing model explanations can be a way of increasing the understanding of “black box” models and building trust. In this work, we applied transfer learning to develop a deep neural network to predict sex from electrocardiograms. Using the visual explanation method Grad-CAM, heat maps were generated from the model in order to understand how it makes predictions. To evaluate the usefulness of the heat maps and determine if the heat maps identified electrocardiogram features that could be recognized to discriminate sex, medical doctors provided feedback. Based on the feedback, we concluded that, in our setting, this mode of explainable artificial intelligence does not provide meaningful information to medical doctors and is not useful in the clinic. Our results indicate that improved explanation techniques that are tailored to medical data should be developed before deep neural networks can be applied in the clinic for diagnostic purposes.