HASTE: Hierarchical Analysis of Spatial and Temporal Data
From intelligent data acquisition via smart data-management to confident predictions
The HASTE project was featured in the June 2022 issue of Framtidens Forskning. It includes the statements from Carolina Wählby, the principal investigator of the project, Salman Toor, senior researcher in the project, and Ida-Maria Sintorn, the industry partner from Vironova AB.
For more information: https://strategiska.se/nu-ar-den-har-tidningen-framtidens-forskning/
Direct link to the pdf: https://framtidensforskning.se/wp-content/uploads/sites/9/2022/06/framtidens-forskning-2022-56s.pdf
Ebba Bergman presented a poster at the 11th Symposium on Pharmaceutical Profiling in Drug Discovery and Development with the title “Conformal Prediction applied to Cell Painting: Confidence in MoA Prediction”. The conference was held online on January 27th, 2022.
We are happy to announce that our paper “Proactive Autoscaling for Edge Computing Systems with Kubernetes” is now accepted at the 14th IEEE/ACM International Conference on Utility and Cloud Computing UCC 2021.
With the emergence of the Internet of Things and 5G technologies, the edge computing paradigm is playing increasingly important roles with better availability, latency-control and performance. However, existing autoscaling tools for edge computing applications do not utilize heterogeneous resources of edge systems efficiently, leaving scope for performance improvement. In this work, we propose a Proactive Pod Autoscaler (PPA) for edge computing applications on Kubernetes. The proposed PPA is able to forecast workloads in advance with multiple user-defined/customized metrics and to scale edge computing applications up and down correspondingly. The PPA is optimized and evaluated on an example CPU-intensive edge computing application further. It can be concluded that the proposed PPA outperforms the default pod autoscaler of Kubernetes on both efficiencies of resource utilization and application performance. The article also highlights future possible improvements on the proposed PPA.
Currently available (arXiv link): https://arxiv.org/pdf/2112.10127.pdf
We are happy to present the joining of Xiaobo Zhao as our newest member of HASTE. Xiaobo Zhao is joining the group of Andreas Hellander to work as a PostDoctoral Researcher. Xiaobo will be working on research and development of intelligent stream data processing pipelines, and the development of intelligent and efficient cloud systems capable of mapping data and compute to a variety of cloud computing and data storage e-infrastructure based on the quality and interestingness of the data.
Xiaobo Zhao received the M.S. degree in Communications and Information System from Northwestern Polytechnical University, Xi’an, China in 2015. He later received a Ph.D. degree in Electrical and Computer Engineering from Aarhus University, Aarhus, Denmark in 2020. Before joining the Hellander lab, he was a Research Assistant at Aarhus University and focused on efficient ML/DL service offloading to Edge/Cloud servers.
We are happy to announce that our paper “Deep learning models for lipid-nanoparticle-based drug delivery” is now available ahead of print and open access in the journal Nanomedicine.
Authors: Harrison PJ, Wieslander H, Sabirsh A, Karlsson J, Malmsjö V, Hellander A, Wählby C & Spjuth O.
DOI: 10.2217/nnm-2020-0461
Abstract:
Background: Early prediction of time-lapse microscopy experiments enables intelligent data management and decision-making. Aim: Using time-lapse data of HepG2 cells exposed to lipid nanoparticles loaded with mRNA for expression of GFP, the authors hypothesized that it is possible to predict in advance whether a cell will express GFP. Methods: The first modeling approach used a convolutional neural network extracting per-cell features at early time points. These features were then combined and explored using either a long short-term memory network (approach 2) or time series feature extraction and gradient boosting machines (approach 3). Results: Accounting for the temporal dynamics significantly improved performance. Conclusion: The results highlight the benefit of accounting for temporal dynamics when studying drug delivery using high-content imaging.
In the figure below we show a schematic for the modelling approach used in the paper that combined convolutional and recurrent neural networks (long short-term memory, LSTM). This model is used for predicting information only present in the GFP channel at the end of the experiment from other imaging channels captured during the early time points of the experiment, prior to any GFP expression.
Following the win at the Adipocyte Imaging Challenge organized by AstraZeneca, two PhD students from the team, Ankit Gupta, and Håkan Wieslander were asked to comment in a technical report in Nature on the topic of virtual staining.
The technology feature can be found here:
Fluorescence imaging is a valuable tool for biological analysis but is time-consuming and toxic to the cells. Using deep learning to virtually stain bright-field images is an active field of research that can alleviate these problems. Phil Harrison, a PhD student in the HASTE group, presented a poster at the Swedish Symposium on Deep Learning (SSDL) 2021 based on the HASTE team’s winning solution for the Adipocyte Cell Imaging Challenge. The poster presented our approach and results.
AI Sweden and AstraZeneca organised the Adipocyte Cell Imaging Challenge, a two-week-long hackathon to help AstraZeneca accelerate the drug development process. The task was to use machine learning in solving the problem of labelling cell images without requiring toxic preprocessing of cell cultures by predicting the content of the fluorescence images from the corresponding bright-field images.
Eight teams consisting of Swedish and international representatives from both academia and the private sector participated in the challenge. Our winning team consisted of Ankit Gupta, Håkan Wieslander, Phil Harrison, Ebba Bergman from HASTE Team and Erik Hallström from Wählby lab. The winner was selected by a jury consisting of representatives from AstraZeneca, Vinnova and AI Sweden.
The solution used the Learning Under Privileged Information (LUPI) paradigm to solve the problem. LUPI enables the inclusion of additional (privileged) information when training machine learning models, data that is not available when making predictions. In this case, the segmentation masks of the nuclei were used as the privileged information during the training of machine learning models. Our solution will help AstraZeneca to speed up the drug discovery process and bring drugs to market quicker.
More information on challenge can be found here.
Press release about the solution can be found here, here, and here.
Unfortunately, we didn’t get a chance to enjoy the good food and beautiful environment of the Noor Castle, but we had a productive project meeting anyway. The meeting started with report reading and Salman Toor presenting his docent lecture “Distributed Computing e-Infrastructures (DCI): Challenges and Opportunities for Application Experts and Service Providers”. This was followed by intense brainstorming on the continued research in the project, and a discussion on the almost complete half-time report, and planning for the coming Tuesday seminars, which have become an important part of the project communication now that most of us work from home.
Hongru is a master’s student from the department of statistics, and his main interests in statistics include multivariate statistical methods and Bayesian statistics.
Hongru’s MSc thesis will focus on developing the better hierarchical representations of the microscopy image data from cellular experiments with the help of statistical methods, focusing on improving readability and informational efficiency of the representation.
