ELLIS against Covid-19

Covid-19 Projects

A mathematical model on the COVID-19 spreading at its first stage


In this work, a new Stochastic process different from the well known Polya model is proposed. The Polya stochastic process arises from the Polya urn model for contagion as a limit and its mean number of successful events is a linear function of time as it is demonstrated. The Polya process can be described as a Pure Birth process with a given events rate that depends over time and over the previous number of events. In our proposal, we suggest to use a Pure Birth process with a slight variation in the functional form of the events rate. From our proposal, we can get a process more or less contagious than the Polya process provided a given parameter is greater or lower than one. Applications to the recent SARS-COVID-19 pandemic is shown, with good agreement. This model allows to predict the contagion evolution at the first stage of the pandemic, when the disease is spreading out before achieving the inflexion point.

Contact: Nestor R. Barraza, Universidad Nacional de Tres de Febrero (nestor.barraza@gmail.com

A Spatiotemporal Epidemic Model to Quantify the Effects of Contact Tracing, Testing, and Containment


We introduce a novel modeling framework for studying epidemics that is specifically designed to make use of fine-grained spatiotemporal data. Motivated by the current COVID-19 outbreak and the availability of data from contact or location tracing technologies, our model uses marked temporal point processes to represent individual mobility patterns and the course of the disease for each individual in a population. We design an efficient sampling algorithm for our model that can be used to predict the spread of infectious diseases such as COVID-19 under different testing and tracing strategies, social distancing measures, and business restrictions, given location or contact histories of individuals. Building on this algorithm, we use Bayesian optimization to estimate the risk of exposure of each individual at the sites they visit, the percentage of symptomatic individuals, and the difference in transmission rate between asymptomatic and symptomatic individuals from historical longitudinal testing data.

Experiments using measured COVID-19 data and mobility patterns from Tübingen, a town in the southwest of Germany, demonstrate that our model can be used to quantify the effects of tracing, testing, and containment strategies at an unprecedented spatiotemporal resolution. To facilitate research and informed policy-making, particularly in the context of the current COVID-19 outbreak, we are releasing an open-source implementation of our framework at https://github.com/covid19-model.


Manuel Gomez Rodriguez, ELLIS unit Saarbrücken

Bernhard Schölkopf, ELLIS unit Tübingen

Project website: https://arxiv.org/abs/2004.07641

AI against COVID-19@Mila

In light of the challenges that COVID-19 presents to our society, Mila is bringing its machine learning expertise to the scientific community together with its partners across different disciplines to help find potential solutions. A list of Covid related projects at Mila, some of them in cooperation with ELLIS members can be found here.

The Mila COVID-19 initiative is part of the AI against COVID-19 Canada Task Force. Here you can find a more comprehensive list of ongoing projects in the Canadian AI Research Institutes together with a curated lists of available datasets and their ML Approach and/or Application, as well as a list of relevant publications about COVID-19 and Machine Learning.

AI@RISC-19-ICU. RIsk Stratification in COVID-19 patients in the ICU


The aim of the research is the design and development of Artificial Intelligence and Machine Learning methods (ML) for the prediction of complications and risk stratification of COVID 19 Intensive Care Units (ICUs) using heterogeneous longitudinal Electronic Health Record (EHR) data. In particular, the study will be performed as part of the "Collaborative international ICU registry for critically ill COVID-19 patients - RISC-19-ICU" (https://sites.google.com/view/risc-19-icu).

The Visual Robotics and Artificial Intelligence (VRAI) group from the Department of Information Engineering, Università Politecnica delle Marche, Ancona, Italy supervised by Prof. Emanuele Frontoni will be responsible to design and develop AI and ML methods, AI data architectures and learning-testing procedures.

The idea behind the project is to better prevent and treat the complication that is about to appear in the patient affected by COVID19 by developing a decision support system that allows computing:

  • risk profiles of individual patients from which a different intensity of care can be deduced, with consequent modification of the control time according to the needs; this approach would produce a shortening of the waiting time and an increase in the appropriateness of care.

  • prediction of risk of short-term complications which will activate personalized prevention systems directly addressed to the patient: from targeted recalls to targeted motivational and training activities

Objectives (AI/ML Section) 

The analysis of EHR data using statistical and Machine Learning methodologies aims to:

  • develop innovative solutions and predictions whether a patient belongs to a risk class related to specific COVID-19 complications over a short period of time.  

  • the extraction of clinical features of EHR (e.g. specific drugs) are related to the development of the complication. 

The developing a clinical decision support system, integrated in an electronic record, that allows the risk profile of morbidity and resource consumption of the patients according to their clinical features would allow more appropriate and intensive use of resources for those who need them most.


Publication 1: Machine learning using the extreme gradient boosting (XGBoost) algorithm predicts 5-day delta of SOFA score at ICU admission in COVID-19 patients

Background: Accurate risk stratification of critically ill patients with coronavirus disease 2019 (COVID-19) is essential for optimizing resource allocation, delivering targeted interventions, and maximizing patient survival probability. Machine learning (ML) techniques are attracting increased interest for the development of prediction models as they excel in the analysis of complex signals in data-rich environments such as critical care. Methods: We retrieved data on patients with COVID-19 admitted to an intensive care unit (ICU) between March and October 2020 from the RIsk Stratification in COVID-19 patients in the Intensive Care Unit (RISC-19-ICU) registry. We applied the Extreme Gradient Boosting (XGBoost) algorithm to the data to predict as a binary out- come the increase or decrease in patients’ Sequential Organ Failure Assessment (SOFA) score on day 5 after ICU admission. The model was iteratively cross-validated in different subsets of the study cohort. Results: The final study population consisted of 675 patients. The XGBoost model correctly predicted a decrease in SOFA score in 320/385 (83%) critically ill COVID-19 patients, and an increase in the score in 210/290 (72%) patients. The area under the mean receiver operating characteristic curve for XGBoost was significantly higher than that for the logistic regression model (0.86 vs. 0.69, P < 0.01 [paired t-test with 95% confidence interval]). Conclusions: The XGBoost model predicted the change in SOFA score in critically ill COVID-19 patients admitted to the ICU and can guide clinical decision support systems (CDSSs) aimed at optimizing available resources.

Link: https://medcentral.net/doi/pdf/10.1016/j.jointm.2021.09.002


Publication 2: Harnessing the Power of Smart and Connected Health to Tackle COVID-19: IoT, AI, Robotics, and Blockchain for a Better World

As COVID-19 hounds the world, the common cause of finding a swift solution to manage the pandemic has brought together researchers, institutions, governments, and society at large. The Internet of Things (IoT), artificial intelligence (AI)-including machine learning (ML) and Big Data analytics-as well as Robotics and Blockchain, are the four decisive areas of technological innovation that have been ingenuity harnessed to fight this pandemic and future ones. While these highly interrelated smart and connected health technologies cannot resolve the pandemic overnight and may not be the only answer to the crisis, they can provide greater insight into the disease and support frontline efforts to prevent and control the pandemic. This article provides a blend of discussions on the contribution of these digital technologies, propose several complementary and multidisciplinary techniques to combat COVID-19, offer opportunities for more holistic studies, and accelerate knowledge acquisition and scientific discoveries in pandemic research. First, four areas, where IoT can contribute are discussed, namely: 1) tracking and tracing; 2) remote patient monitoring (RPM) by wearable IoT (WIoT); 3) personal digital twins (PDTs); and 4) real-life use case: ICT/IoT solution in South Korea. Second, the role and novel applications of AI are explained, namely: 1) diagnosis and prognosis; 2) risk prediction; 3) vaccine and drug development; 4) research data set; 5) early warnings and alerts; 6) social control and fake news detection; and 7) communication and chatbot. Third, the main uses of robotics and drone technology are analyzed, including: 1) crowd surveillance; 2) public announcements; 3) screening and diagnosis; and 4) essential supply delivery. Finally, we discuss how distributed ledger technologies (DLTs), of which blockchain is a common example, can be combined with other technologies for tackling COVID-19

Link: https://ieeexplore.ieee.org/abstract/document/9406879


Publication 3: A Unified Hierarchical XGBoost model for classifying priorities for COVID-19 vaccination campaign

The current ML approaches do not fully focus to answer a still unresolved and topical challenge, namely the prediction of priorities of COVID-19 vaccine administration. Thus, our task includes some additional methodological challenges mainly related to avoiding unwanted bias while handling categorical and ordinal data with a highly imbalanced nature. Hence, the main contribution of this study is to propose a machine learning algorithm, namely Hierarchical Priority Classification eXtreme Gradient Boosting for priority classification for COVID-19 vaccine administration using the Italian Federation of General Practitioners dataset that contains Electronic Health Record data of 17k patients. We measured the effectiveness of the proposed methodology for classifying all the priority classes while demonstrating a significant improvement with respect to the state of the art. The proposed ML approach, which is integrated into a clinical decision support system, is currently supporting General Pracitioners in assigning COVID-19 vaccine administration priorities to their assistants.

Link: https://www.sciencedirect.com/science/article/pii/S0031320321003794


The RISC-19-ICU board includes Dr. M. P. Hilty, P. D. Wendel Garcia, MSc, Prof. Dr. R. Schuepbach, Dr. J. Montomoli, Dr. Ph. Guerci, Prof. Dr. T. Fumeaux. The aim of the registry is to collect real-time data of COVID patients admitted to non-ICU or ICU wards. The registry was launched on the 13th of March and it includes already 75 centers from 14 countries collecting data.

Contact: Luca Romeo (l.romeo@staff.univpm.it)

Covid -19 related projects at the ELLIS unit Genoa

Project 1: AiThermometer

Open source project for automatically measuring the temperature of people using a thermal camera. The software can be freely used for any non-commercial applications and it is useful for the automatic early-screening of fever symptoms. The software first detect people with an off-the-shelf body pose detector and then extract location of the face where the temperature is measured. The system requires a known reference temperature and the value and position are provided by the user (this information is shown as a single small green circle on the image). It is possible to have the absolute temperature but it requires an image from a thermal camera with correct radiometric calibration and radiometric exif data loaded into image. https://github.com/IIT-PAVIS/AI-Thermometer

Project 2: Radiomics for COVID-19 early screening

The main aim of this task is to develop radiomics tools to analyze thoracic X-ray and CT exams, in order to diagnose interstitial pneumonia and predict a likely prognosis in collaboration with Bracco imaging. This would allow physicians to anticipate the severity of the disease and, therefore, select the best clinical management of the patient. The goal is to realize a flexible architecture generalizing across multiple centers and modalities according to the adopted clinical protocol, which could also be quickly modified and deployed in the current and future health crises.

Project 3: Molecular simulation

We have examined a series of molecular targets for COVID-19 and performed in silico screening of a database containing more than 3,000 FDA-authorized drugs. We will use docking, virtual screening, and machine learning to identify and develop new small molecule hits and to repurpose existing drugs (a much faster process compared to new drug development).

Contact: Alessio del Blue (alessio.delbue@iit.it

COVID-19 Identification, Estimating the Severity, and Dynamic Evolution Prediction via Explainable Deep Learning

The aim of this research is to study and develop efficient yet explainable methods to identify COVID-19 via chest X-ray and thoracic computed tomography (CT). While manual reading CT and X-ray images takes 15 minutes and involves a highly skilled medical doctor/consultant which are now in high demand, the use of machine learning can take few seconds on a computer and be automated which provides opportunity for high throughput and remote way of operation. 

In this research proposal we suggest the use of a new classification method that is explainable by design and able to continue to learn and adapt for each new data sample. The latter is immensely important for the case of COVID-19 (and other disease), because new cases are being accumulated every minute and traditional approaches require either iterative re-training or ignore the new data. We aim to provide an understandable and interpretable recommendation framework that can be important decision support tool to specialists in the diagnosis and treatment decision-making process. 

To train our model we plan to use open source data, as well as, provide our own dataset. The data that we aim to provide is being collected from different hospitals of Sao Paulo,Brazil. We expect to acquire data for 560 patients divided between 280 patients with COVID-19 and 280 non-COVID patients for control. These data will be constituted of thoracic CT images (at least 10 different images per patient) and qualitative data such as age, profession, sex, habits, previous diseases, and others. 

The main idea is to provide i) an explainable COVID-19 identification tool via thoracic CT images; ii) a tool to estimate the degree of pulmonary involvement by COVID-19; iii) and a tool predict the risk of dynamic evolution of the pulmonary involvement via data fusion of qualitative features and thoracic CT. In the third case, the explainable deep learning method will be trained having as input the qualitative data provided by a patient and thoracic CT time series (for patients that have done more than one CT-scan in different days) in order to predict the degree of pulmonary involvement for a given n number of days.


Plamen Angelov (p.angelov@lancaster.ac.uk)


COVID-19 Open Research Dataset (CORD-19)

In response to the COVID-19 pandemic, the Allen Institute for AI has released the COVID-19 Open Research Dataset (CORD-19), a free resource of over 45,000 scholarly articles, including over 33,000 with machine-readable full text, about COVID-19 and the coronavirus family of viruses for use by the global research community.

This dataset is intended to mobilize researchers to apply recent advances in natural language processing to generate new insights in support of the fight against this infectious disease. The corpus is being updated weekly as new research is published in peer-reviewed publications and archival services like bioRxiv, medRxiv, and others.

Link:  https://pages.semanticscholar.org/coronavirus-research

Discussion: https://discourse.cord-19.semanticscholar.org/


This is in collaboration with the White House OSTP, the National Library of Medicine (NLM), the Chan Zuckerberg Initiative (CZI), Microsoft Research (MSR), and Georgetown University's Center for Science and Emerging Technology (CSET).

Contact: Kyle Lo (kylel@allenai.org)

Project website: https://pages.semanticscholar.org/coronavirus-research

COVID-19 Vaccine Candidates: Prediction and Validation of 174 SARS-CoV-2 Epitopes

The recent outbreak of SARS-CoV-2 (2019-nCoV) virus has highlighted the need for fast and efficacious vaccine development. Stimulation of a proper immune response that leads to protection is highly dependent on presentation of epitopes to circulating T-cells via the HLA complex. SARS-CoV-2 is a large RNA virus and testing of all overlapping peptides ​in vitro​ to deconvolute an immune response is not feasible. Therefore HLA-binding prediction tools are often used to narrow down the number of peptides to test. We tested 15 epitope-HLA-binding prediction tools, and using an ​in vitro​ peptide MHC stability assay, we assessed 777 peptides that were predicted to be good binders across 11 MHC allotypes. In this investigation of potential SARS-CoV-2 epitopes we found that current prediction tools vary in performance when assessing binding stability, and they are highly dependent on the MHC allotype in question. Designing a COVID-19 vaccine where only a few epitope targets are included is therefore a very challenging task. Here, we present 174 SARS-CoV-2 epitopes with high prediction binding scores, validated to bind stably to 11 HLA allotypes. Our findings may contribute to the design of an efficacious vaccine against COVID-19.



Please see complete list of authors here: 


Contact: Frederik Otzen Bagger (frederik.otzen.bagger@regionh.dk)

COVIDSearch. Making Sense of [Lots of] Open Data Related to COVID-19


The overall goals of this challenge evaluation are to:

  • Evaluate search algorithms and systems for helping scientists, clinicians, policy makers, and others manage the existing and the rapidly growing corpus of scientific literature related to COVID-19

  • Discover methods that will assist with managing scientific information in future global biomedical crises




Contact: Djoerd Hiemstra (Djoerd.Hiemstra@ru.nl

Project website: https://dmice.ohsu.edu/hersh/COVIDSearch.html

Fast detection of SARS-CoV-2 contaminated surfaces using multispectral holographic and optical analysis with artificial intelligence

Currently, there are no methods to detect and visualize the presence of the SARS-CoV2 virus on material surfaces. It is proposed the design, development and evaluation of a portable prototype device (using existing optical and photonics technologies combined with machine learning/artificial intelligence) for the rapid and contactless detection, in-situ, of contaminated areas on surfaces. We propose to i) combine multispectral images in the optical (ultraviolet to thermal infrared) and terahertz ranges, ii) use methods of analysis by computational optics, interferometry and holography, iii) integrate the information using artificial intelligence, and iv) perform tests in laboratory, and in clean and contaminated environments.


  • Universidad de Sevilla, Spain (coordinator)
  • EOD-CBRN Group of the Spanish National Police
  • University Hospital ‘Virgen del Rocío’ (Seville, Spain)
  • University Hospital ‘Virgen Macarena’ (Seville, Spain)
  • Institute of Biomedicine of Seville (IBIS), Spain
  • Andalusian Network of Design and Translation of Advanced Therapies (RAdytTA), Spain
  • University of Cádiz-INIBICA, Spain
  • Astronomical Observatory ‘Calar Alto’ (CAHA, Almería, Spain)
  • Institute of Astrophysics of Andalucia (IAA)-CSIC (Granada, Spain)
  • Technological Corporation of Andalusia (CTA), Spain
  • with the collaboration of the HUMAINT Project of the Joint Research Center (JRC), European Commission (EC). 

Results: www.nature.com/articles/s41598-021-95756-3

Project website: en.gfi-us.org/proyecto-c-clean


Contact: Emilio Gomez-Gonzalez (Universidad de Sevilla, Spain) – egomez@us.es



The team of Spanish researchers coordinated by the University of Seville has published the first results ('proof of concept') of detecting the coronavirus that causes COVID-19 using a new optical methodology. This tool could be potentially usable for massive, fast and easy-to implement screening. This new methodology, whose first results are published in the journal Scientific Reports from the Nature Group has obtained a sensitivity of 100% and a specificity of 87.5% in the detection of SARS-CoV-2 in nasopharyngeal exudate (the same samples used in a PCR test) from symptomatic people. It has also been possible to detect the presence of SARS-CoV-2 in fresh saliva of asymptomatic people, as well as to detect, differentiate and quantify two types of synthetic viruses (lentiviruses and synthetic coronaviruses) in two biofluids (saline solution and artificial saliva). The main advantage of this new technology over PCR lies in the speed of sample processing and the ability of the optical system to simultaneously analyze a large number of samples.

Find out more in this press release:

Health empowerment through smartphone sensing

The aims are described here https://coronamonitor.dk/  (in Danish).

Briefly summarized: the team will develop a suite of tools for self-monitoring of potential covid19 symptoms, based on smartphone sensing. The first service is based on cough detection and classification with DNNs. All tools and models will be open source. The interdisciplinary team involves expertise in bio-medicine, UX, software eng, and machine learning.


Interested in collaboration on:

*Data and ML workflows

*UX incl privacy designscheers,

Contact: Lars Kai Hansen (lars.kai.hansen@gmail.com)

Mobile phone data and COVID-19: Missing an opportunity?

This study describes how mobile phone data can guide government and public health authorities in determining the best course of action to control the COVID-19 pandemic and in assessing the effectiveness of control measures such as physical distancing. It identifies key gaps and reasons why this kind of data is only scarcely used, although their value in similar epidemics has proven in a number of use cases. It presents ways to overcome these gaps and key recommendations for urgent action, most notably the establishment of mixed expert groups on national and regional level, and the inclusion and support of governments and public authorities early on. 

It is authored by a group of experienced data scientists, epidemiologists, demographers and representatives of mobile network operators who jointly put their work at the service of the global effort to combat the COVID-19 pandemic. 



Bruno Lepri (lepri@fbk.eu

Nuria Oliver (nuriaoliver@outlook.com)

Please see complete list of authors here: https://arxiv.org/ftp/arxiv/papers/2003/2003.12347.pdf

Contact: Patrick Vinck (pvinck@hsph.harvard.edu)

Post-first-wave Covid 19 exit strategies

The initial UK government strategy of a timed intervention as a means of combatting Covid-19 is in stark contrast to policies adopted in many other countries which have embraced more severe social distancing policies. Our objective in this note is to suggest modified policies, for post lockdown, that may allow management of Covid-19, while at the same time enabling some degree of reduced social and economic activity.

A brief report can be found at:


Contact person: Robert Shorten (r.shorten@imperial.ac.uk)

Project website: https://robertshorten.files.wordpress.com/2020/03/fpsr_title_version_1_4.pdf

Predicting antigen-specificity of single T-cells based on TCR CDR3 regions

It has recently become possible to assay T-cell specificity with respect to large sets of antigens as well as T-cell receptor sequence in high-throughput single-cell experiments. We propose multiple sequence-data specific deep learning approaches to impute TCR to epitope specificity to reduce the complexity of new experiments. We found that models that treat antigens as categorical variables outperform those which model the TCR and epitope sequence jointly. Moreover, we show that variability in single-cell immune repertoire screens can be mitigated by modeling cell-specific covariates.

Please see complete list of authors here: 


Contact: Fabian Theis (fabian.theis@helmholtz-muenchen.de)

Project website: https://www.biorxiv.org/content/10.1101/734053v1

Resource optimising real-time analysis of artifactious image sequences for the detection of nano-objects


The comprehensive real-time detection of the Coronavirus SARS-CoV-2 is a fundamental challenge. The biosensor called PAMONO (for Plasmon Assisted Microscopy of Nano-sized Objects) could make a valuable contribution here. The sensor represents a viable technology for mobile real-time detection and quantitative analysis of viruses and virus-like particles. A mobile system that can detect viruses in real-time is urgently needed, due to the combination of virus emergence and evolution with increasing global travel and transport. It could be used for fast and reliable diagnoses in hospitals, airports, the open air, or other settings. The development of the sensor is part of the collaborative research center 876 funded by DFG (sfb876.tu-dortmund.de) and has been launched since 2010.

The PAMONO-sensor permits the imaging of biological nano-vesicles (e.g. the Coronavirus) utilizing a Kretschmann’s scheme of plasmon excitation with an illumination of a gold sensor surface via a glass prism. The PAMONO sensor applies anti-bodies to bind the nano-sized viruses on a gold layer. The presence of viruses can be detected by the intensity change of the reflection of a laser beam. Characteristics of these binding events are spatiotemporal blob-like structures with very low signal-to-noise ratio, which indicate particle bindings and can be automatically analyzed with image processing methods. We capture the intensity of the reflected laser beams using a CCD camera, which result in a series of artifactious images.

For the analysis of the images provided by the sensor, we have developed nanoparticle classification approaches based on deep neural network architectures. It is shown that the combination of the PAMONO sensor and the application of deep learning enables a real-time data processing to automatically detect and quantify biological particles. With the availability of anti-SARS-CoV-2 antibodies, the PAMONO-sensor could thus also be trained to detect the Coronavirus. For more technical details, we refer the reader to our survey paper by Shpacovitch, et al. http://dx.doi.org/10.3390/s17020244



Dr. Roland Hergenröder (Leibniz Institute for Analytical Sciences, Dortmund) 


Email: roland.hergenroeder@isas.de

Priv. Doz. Dr. Frank Weichert (Faculty Computer Science, TU Dortmund University)


Email: frank.weichert@tu-dortmund.de

Prof. Jian-Jia Chen (Faculty Computer Science, TU Dortmund University)


Email: jian-jia.chen@tu-dortmund.de


Contact: Prof. Katharina Morik (https://sfb876.tu-dortmund.de)

SARS-CoV-2 receptor ACE2 is an interferon-stimulated gene in human airway epithelial cells and is detected in specific cell subsets across tissues

There is pressing urgency to understand the pathogenesis of the severe acute respiratory syndrome coronavirus clade 2 (SARS-CoV-2) which causes the disease COVID-19. SARS-CoV- 2 spike (S)-protein binds ACE2, and in concert with host proteases, principally TMPRSS2, promotes cellular entry. The cell subsets targeted by SARS-CoV-2 in host tissues, and the factors that regulate ACE2 expression, remain unknown. Here, we leverage human, non-human primate, and mouse single-cell RNA-sequencing (scRNA-seq) datasets across health and disease to uncover putative targets of SARS-CoV-2 amongst tissue-resident cell subsets. We identify ACE2 and TMPRSS2 co-expressing cells within lung type II pneumocytes, ileal absorptive enterocytes, and nasal goblet secretory cells. Strikingly, we discover that ACE2 is a human interferon- stimulated gene (ISG) in vitro using airway epithelial cells, and extend our findings to in vivo viral infections. Our data suggest that SARS-CoV-2 could exploit species-specific interferon-driven upregulation of ACE2, a tissue-protective mediator during lung injury, to enhance infection.

Contact: Fabian Theis (fabian.theis@helmholtz-muenchen.de)

Project website: https://www.cell.com/pb-assets/products/coronavirus/CELL_CELL-D-20-00767.pdf

Support your NHS


The project is a data capture tool aimed at collecting information about the number of people who are in isolation in the UK, due to Covid-19. The collected data is presented and the goal is to assist the National Health Service's trusts to understand how this pandemic is driving forward, but also to help  the citizens of the UK to understand how both the area they live in, and the areas that their family and loved ones reside are affected by this pandemic.








Keitaro (https://www.keitaro.com)

Nosy Design (https://www.nosydesign.co.uk

Mapsimise (https://mapsimise.com)



Marina Tanevska (marina.tanevska@keitaro.com)