Search

Your search keyword '"Kringen, Pedro"' showing total 20 results
Start Over Author "Kringen, Pedro"

Refine your results

more »

more »

more »
20 results on '"Kringen, Pedro"'

1. How to Assess Trustworthy AI in Practice

Author

Zicari, Roberto V., Amann, Julia, Bruneault, Frédérick, Coffee, Megan, Düdder, Boris, Hickman, Eleanore, Gallucci, Alessio, Gilbert, Thomas Krendl, Hagendorff, Thilo, van Halem, Irmhild, Hildt, Elisabeth, Holm, Sune, Kararigas, Georgios, Kringen, Pedro, Madai, Vince I., Mathez, Emilie Wiinblad, Tithi, Jesmin Jahan, Vetter, Dennis, Westerlund, Magnus, and Wurth, Renee

Subjects
Computer Science - Computers and Society
Abstract

This report is a methodological reflection on Z-Inspection$^{\small{\circledR}}$. Z-Inspection$^{\small{\circledR}}$ is a holistic process used to evaluate the trustworthiness of AI-based technologies at different stages of the AI lifecycle. It focuses, in particular, on the identification and discussion of ethical issues and tensions through the elaboration of socio-technical scenarios. It uses the general European Union's High-Level Expert Group's (EU HLEG) guidelines for trustworthy AI. This report illustrates for both AI researchers and AI practitioners how the EU HLEG guidelines for trustworthy AI can be applied in practice. We share the lessons learned from conducting a series of independent assessments to evaluate the trustworthiness of AI systems in healthcare. We also share key recommendations and practical suggestions on how to ensure a rigorous trustworthy AI assessment throughout the life-cycle of an AI system., Comment: On behalf of the Z-Inspection$^{\small{\circledR}}$ initiative (2022)

Published
2022
Full Text
View/download PDF

2. Lessons Learned from Assessing Trustworthy AI in Practice

Author

Vetter, Dennis, Amann, Julia, Bruneault, Frédérick, Coffee, Megan, Düdder, Boris, Gallucci, Alessio, Gilbert, Thomas Krendl, Hagendorff, Thilo, Van Halem, Irmhild, Hickman, Eleanore, Hildt, Elisabeth, Holm, Sune, Kararigas, Georgios, Kringen, Pedro, Madai, Vince I., Wiinblad Mathez, Emilie, Tithi, Jesmin Jahan, Westerlund, Magnus, Wurth, Renee, Zicari, Roberto V., Vetter, Dennis, Amann, Julia, Bruneault, Frédérick, Coffee, Megan, Düdder, Boris, Gallucci, Alessio, Gilbert, Thomas Krendl, Hagendorff, Thilo, Van Halem, Irmhild, Hickman, Eleanore, Hildt, Elisabeth, Holm, Sune, Kararigas, Georgios, Kringen, Pedro, Madai, Vince I., Wiinblad Mathez, Emilie, Tithi, Jesmin Jahan, Westerlund, Magnus, Wurth, Renee, and Zicari, Roberto V.

Abstract

Building artificial intelligence (AI) systems that adhere to ethical standards is a complex problem. Even though a multitude of guidelines for the design and development of such trustworthy AI systems exist, these guidelines focus on high-level and abstract requirements for AI systems, and it is often very difficult to assess if a specific system fulfills these requirements. The Z-Inspection® process provides a holistic and dynamic framework to evaluate the trustworthiness of specific AI systems at different stages of the AI lifecycle, including intended use, design, and development. It focuses, in particular, on the discussion and identification of ethical issues and tensions through the analysis of socio-technical scenarios and a requirement-based framework for ethical and trustworthy AI. This article is a methodological reflection on the Z-Inspection® process. We illustrate how high-level guidelines for ethical and trustworthy AI can be applied in practice and provide insights for both AI researchers and AI practitioners. We share the lessons learned from conducting a series of independent assessments to evaluate the trustworthiness of real-world AI systems, as well as key recommendations and practical suggestions on how to ensure a rigorous trustworthiness assessment throughout the lifecycle of an AI system. The results presented in this article are based on our assessments of AI systems in the healthcare sector and environmental monitoring, where we used the framework for trustworthy AI proposed in the Ethics Guidelines for Trustworthy AI by the European Commission’s High-Level Expert Group on AI. However, the assessment process and the lessons learned can be adapted to other domains and include additional frameworks.

Published
2023

3. Co-Design of a Trustworthy AI System in Healthcare: Deep Learning Based Skin Lesion Classifier

Author

Zicari, Roberto V., Ahmed, Sheraz, Amann, Julia, Braun, Stephan A., Brodersen, John, Bruneault, Frédérick, Brusseau, James, Campano, Erik, Coffee, Megan, Dengel, Andreas, Düdder, Boris, Gallucci, Alessio, Krendl Gilbert, Thomas, Gottfrois, Philippe, Goffi, Emmanuel, Bjerre Haase, Christoffer, Hagendorff, Thilo, Hickman, Eleanore, Hildt, Elisabeth, Holm, Sune, Kringen, Pedro, Kühne, Ulrich, Lucieri, Adriano, Madai, Vince I., Moreno-Sánchez, Pedro A., Medlicott, Oriana, Ozols, Matiss, Schnebel, Eberhard, Spezzatti, Andy, Tithi, Jesmin J., Umbrello, Steven, Vetter, Dennis, Volland, Holger, Westerlund, Magnus, Wurth, Renee, Apollo - University of Cambridge Repository, Security, and Eindhoven MedTech Innovation Center

Subjects
Z-inspection®1, Artificial intelligence, Healthcare, Trustworthy AI, Ethics, Malignant melanoma, Z-inspection ®1, ethical co-design, trustworthy AI Co-design, malignant melanoma, healthcare, Social Sciences, artificial intelligence, ethics, Basic Behavioral and Social Science, 46 Information and Computing Sciences, trustworthy AI, Behavioral and Social Science, ddc:610, ddc:004, 33 Built Environment and Design, 3303 Design
Abstract

This paper documents how an ethically aligned co-design methodology ensures trustworthiness in the early design phase of an artificial intelligence (AI) system component for healthcare. The system explains decisions made by deep learning networks analyzing images of skin lesions. The co-design of trustworthy AI developed here used a holistic approach rather than a static ethical checklist and required a multidisciplinary team of experts working with the AI designers and their managers. Ethical, legal, and technical issues potentially arising from the future use of the AI system were investigated. This paper is a first report on co-designing in the early design phase. Our results can also serve as guidance for other early-phase AI-similar tool developments., Frontiers in Human Dynamics, 3, ISSN:2673-2726

Published
2021
Full Text
View/download PDF

4. On assessing trustworthy AI in healthcare:Best practice for machine learning as a supportive tool to recognize cardiac arrest in emergency calls

Author

Zicari, Roberto V., Brusseau, James, Blomberg, Stig Nikolaj, Christensen, Helle Collatz, Coffee, Megan, Ganapini, Marianna B., Gerke, Sara, Gilbert, Thomas Krendl, Hickman, Eleanore, Hildt, Elisabeth, Holm, Sune, Kühne, Ulrich, Madai, Vince I., Osika, Walter, Spezzatti, Andy, Schnebel, Eberhard, Tithi, Jesmin Jahan, Vetter, Dennis, Westerlund, Magnus, Wurth, Renee, Amann, Julia, Antun, Vegard, Beretta, Valentina, Bruneault, Frédérick, Campano, Erik, Düdder, Boris, Gallucci, Alessio, Goffi, Emmanuel, Haase, Christoffer Bjerre, Hagendorff, Thilo, Kringen, Pedro, Möslein, Florian, Ottenheimer, Davi, Ozols, Matiss, Palazzani, Laura, Petrin, Martin, Tafur, Karin, Tørresen, Jim, Volland, Holger, and Kararigas, Georgios

Subjects
ComputingMethodologies_PATTERNRECOGNITION, GeneralLiterature_MISCELLANEOUS
Abstract

Artificial Intelligence (AI) has the potential to greatly improve the delivery of healthcare and other services that advance population health and wellbeing. However, the use of AI in healthcare also brings potential risks that may cause unintended harm. To guide future developments in AI, the High-Level Expert Group on AI set up by the European Commission (EC), recently published ethics guidelines for what it terms “trustworthy” AI. These guidelines are aimed at a variety of stakeholders, especially guiding practitioners toward more ethical and more robust applications of AI. In line with efforts of the EC, AI ethics scholarship focuses increasingly on converting abstract principles into actionable recommendations. However, the interpretation, relevance, and implementation of trustworthy AI depend on the domain and the context in which the AI system is used. The main contribution of this paper is to demonstrate how to use the general AI HLEG trustworthy AI guidelines in practice in the healthcare domain. To this end, we present a best practice of assessing the use of machine learning as a supportive tool to recognize cardiac arrest in emergency calls. The AI system under assessment is currently in use in the city of Copenhagen in Denmark. The assessment is accomplished by an independent team composed of philosophers, policy makers, social scientists, technical, legal, and medical experts. By leveraging an interdisciplinary team, we aim to expose the complex trade-offs and the necessity for such thorough human review when tackling socio-technical applications of AI in healthcare. For the assessment, we use a process to assess trustworthy AI, called 1Z-Inspection® to identify specific challenges and potential ethical trade-offs when we consider AI in practice.

Published
2021
Full Text
View/download PDF

5. On Assessing Trustworthy AI in Healthcare. Machine Learning as a Supportive Tool to Recognize Cardiac Arrest in Emergency Calls

Author

Zicari, Roberto V., primary, Brusseau, James, additional, Blomberg, Stig Nikolaj, additional, Christensen, Helle Collatz, additional, Coffee, Megan, additional, Ganapini, Marianna B., additional, Gerke, Sara, additional, Gilbert, Thomas Krendl, additional, Hickman, Eleanore, additional, Hildt, Elisabeth, additional, Holm, Sune, additional, Kühne, Ulrich, additional, Madai, Vince I., additional, Osika, Walter, additional, Spezzatti, Andy, additional, Schnebel, Eberhard, additional, Tithi, Jesmin Jahan, additional, Vetter, Dennis, additional, Westerlund, Magnus, additional, Wurth, Renee, additional, Amann, Julia, additional, Antun, Vegard, additional, Beretta, Valentina, additional, Bruneault, Frédérick, additional, Campano, Erik, additional, Düdder, Boris, additional, Gallucci, Alessio, additional, Goffi, Emmanuel, additional, Haase, Christoffer Bjerre, additional, Hagendorff, Thilo, additional, Kringen, Pedro, additional, Möslein, Florian, additional, Ottenheimer, Davi, additional, Ozols, Matiss, additional, Palazzani, Laura, additional, Petrin, Martin, additional, Tafur, Karin, additional, Tørresen, Jim, additional, Volland, Holger, additional, and Kararigas, Georgios, additional

Published
2021
Full Text
View/download PDF

6. Z-Inspection®: A Process to Assess Trustworthy AI

Author

Zicari, Roberto V., primary, Brodersen, John, additional, Brusseau, James, additional, Dudder, Boris, additional, Eichhorn, Timo, additional, Ivanov, Todor, additional, Kararigas, Georgios, additional, Kringen, Pedro, additional, McCullough, Melissa, additional, Moslein, Florian, additional, Mushtaq, Naveed, additional, Roig, Gemma, additional, Sturtz, Norman, additional, Tolle, Karsten, additional, Tithi, Jesmin Jahan, additional, van Halem, Irmhild, additional, and Westerlund, Magnus, additional

Published
2021
Full Text
View/download PDF

7. On assessing trustworthy AI in healthcare : Machine learning as a supportive tool to recognize cardiac arrest in emergency calls

Author

Zicari, Roberto V., Brusseau, James, Blomberg, Stig Nikolaj, Christensen, Helle Collatz, Coffee, Megan, Ganapini, Marianna B., Gerke, Sara, Gilbert, Thomas Krendl, Hickman, Eleanore, Hildt, Elisabeth, Holm, Sune, Kühne, Ulrich, Madai, Vince I., Osika, Walter, Spezzatti, Andy, Schnebel, Eberhard, Tithi, Jesmin Jahan, Vetter, Dennis, Westerlund, Magnus, Wurth, Renee, Amann, Julia, Antun, Vegard, Beretta, Valentina, Bruneault, Frédérick, Campano, Erik, Düdder, Boris, Gallucci, Alessio, Goffi, Emmanuel, Haase, Christoffer Bjerre, Hagendorff, Thilo, Kringen, Pedro, Möslein, Florian, Ottenheimer, Davi, Ozols, Matiss, Palazzani, Laura, Petrin, Martin, Tafur, Karin, Tørresen, Jim, Volland, Holger, Kararigas, Georgios, Zicari, Roberto V., Brusseau, James, Blomberg, Stig Nikolaj, Christensen, Helle Collatz, Coffee, Megan, Ganapini, Marianna B., Gerke, Sara, Gilbert, Thomas Krendl, Hickman, Eleanore, Hildt, Elisabeth, Holm, Sune, Kühne, Ulrich, Madai, Vince I., Osika, Walter, Spezzatti, Andy, Schnebel, Eberhard, Tithi, Jesmin Jahan, Vetter, Dennis, Westerlund, Magnus, Wurth, Renee, Amann, Julia, Antun, Vegard, Beretta, Valentina, Bruneault, Frédérick, Campano, Erik, Düdder, Boris, Gallucci, Alessio, Goffi, Emmanuel, Haase, Christoffer Bjerre, Hagendorff, Thilo, Kringen, Pedro, Möslein, Florian, Ottenheimer, Davi, Ozols, Matiss, Palazzani, Laura, Petrin, Martin, Tafur, Karin, Tørresen, Jim, Volland, Holger, and Kararigas, Georgios

Abstract

Artificial Intelligence (AI) has the potential to greatly improve the delivery of healthcare and other services that advance population health and wellbeing. However, the use of AI in healthcare also brings potential risks that may cause unintended harm. To guide future developments in AI, the High-Level Expert Group on AI set up by the European Commission (EC), recently published ethics guidelines for what it terms “trustworthy” AI. These guidelines are aimed at a variety of stakeholders, especially guiding practitioners toward more ethical and more robust applications of AI. In line with efforts of the EC, AI ethics scholarship focuses increasingly on converting abstract principles into actionable recommendations. However, the interpretation, relevance, and implementation of trustworthy AI depend on the domain and the context in which the AI system is used. The main contribution of this paper is to demonstrate how to use the general AI HLEG trustworthy AI guidelines in practice in the healthcare domain. To this end, we present a best practice of assessing the use of machine learning as a supportive tool to recognize cardiac arrest in emergency calls. The AI system under assessment is currently in use in the city of Copenhagen in Denmark. The assessment is accomplished by an independent team composed of philosophers, policy makers, social scientists, technical, legal, and medical experts. By leveraging an interdisciplinary team, we aim to expose the complex trade-offs and the necessity for such thorough human review when tackling socio-technical applications of AI in healthcare. For the assessment, we use a process to assess trustworthy AI, called 1Z-Inspection® to identify specific challenges and potential ethical trade-offs when we consider AI in practice.

Published
2021
Full Text
View/download PDF

8. Z-Inspection®: A Process to Assess Trustworthy AI

Author

Zicari, Roberto V., Brodersen, John, Brusseau, James, Dudder, Boris, Eichhorn, Timo, Ivanov, Todor, Kararigas, Georgios, Kringen, Pedro, McCullough, Melissa, Moslein, Florian, Mushtaq, Naveed, Roig, Gemma, Sturtz, Norman, Tolle, Karsten, Tithi, Jesmin Jahan, Halem, Irmhild van, Westerlund, Magnus, Zicari, Roberto V., Brodersen, John, Brusseau, James, Dudder, Boris, Eichhorn, Timo, Ivanov, Todor, Kararigas, Georgios, Kringen, Pedro, McCullough, Melissa, Moslein, Florian, Mushtaq, Naveed, Roig, Gemma, Sturtz, Norman, Tolle, Karsten, Tithi, Jesmin Jahan, Halem, Irmhild van, and Westerlund, Magnus

Published
2021

9. Co-design of a trustworthy AI system in healthcare:Deep learning based skin lesion classifier

Author

Zicari, Roberto V., Ahmed, Sheraz, Amann, Julia, Braun, Stephan Alexander, Brodersen, John, Bruneault, Frédérick, Brusseau, James, Campano, Erik, Coffee, Megan, Dengel, Andreas, Düdder, Boris, Gallucci, Alessio, Gilbert, Thomas Krendl, Gottfrois, Philippe, Goffi, Emmanuel, Haase, Christoffer Bjerre, Hagendorff, Thilo, Hickman, Eleanore, Hildt, Elisabeth, Holm, Sune, Kringen, Pedro, Kühne, Ulrich, Lucieri, Adriano, Madai, Vince I., Moreno-Sánchez, Pedro A., Medlicott, Oriana, Ozols, Matiss, Schnebel, Eberhard, Spezzatti, Andy, Tithi, Jesmin Jahan, Umbrello, Steven, Vetter, Dennis, Volland, Holger, Westerlund, Magnus, Wurth, Renee, Zicari, Roberto V., Ahmed, Sheraz, Amann, Julia, Braun, Stephan Alexander, Brodersen, John, Bruneault, Frédérick, Brusseau, James, Campano, Erik, Coffee, Megan, Dengel, Andreas, Düdder, Boris, Gallucci, Alessio, Gilbert, Thomas Krendl, Gottfrois, Philippe, Goffi, Emmanuel, Haase, Christoffer Bjerre, Hagendorff, Thilo, Hickman, Eleanore, Hildt, Elisabeth, Holm, Sune, Kringen, Pedro, Kühne, Ulrich, Lucieri, Adriano, Madai, Vince I., Moreno-Sánchez, Pedro A., Medlicott, Oriana, Ozols, Matiss, Schnebel, Eberhard, Spezzatti, Andy, Tithi, Jesmin Jahan, Umbrello, Steven, Vetter, Dennis, Volland, Holger, Westerlund, Magnus, and Wurth, Renee

Abstract

This paper documents how an ethically aligned co-design methodology ensures trustworthiness in the early design phase of an artificial intelligence (AI) system component for healthcare. The system explains decisions made by deep learning networks analyzing images of skin lesions. The co-design of trustworthy AI developed here used a holistic approach rather than a static ethical checklist and required a multidisciplinary team of experts working with the AI designers and their managers. Ethical, legal, and technical issues potentially arising from the future use of the AI system were investigated. This paper is a first report on co-designing in the early design phase. Our results can also serve as guidance for other early-phase AI-similar tool developments.

Published
2021

10. TP53 mutations in ovarian carcinomas from sporadic cases and carriers of two distinct BRCA1 founder mutations; relation to age at diagnosis and survival

Author

Nesland Jahn M, Dumeaux Vanessa, Wang Yun, Kringen Pedro, Kristensen Gunnar, Borresen-Dale Anne-Lise, and Dorum Anne

Subjects
Neoplasms. Tumors. Oncology. Including cancer and carcinogens, RC254-282
Abstract

Abstract Background Ovarian carcinomas from 30 BRCA1 germ-line carriers of two distinct high penetrant founder mutations, 20 carrying the 1675delA and 10 the 1135insA, and 100 sporadic cases were characterized for somatic mutations in the TP53 gene. We analyzed differences in relation to BRCA1 germline status, TP53 status, survival and age at diagnosis, as previous studies have not been conclusive. Methods DNA was extracted from paraffin embedded formalin fixed tissues for the familial cases, and from fresh frozen specimen from the sporadic cases. All cases were treated at our hospital according to protocol. Mutation analyses of exon 2 – 11 were performed using TTGE, followed by sequencing. Results Survival rates for BRCA1-familial cases with TP53 mutations were not significantly lower than for familial cases without TP53 mutations (p = 0.25, RR = 1.64, 95% CI [0.71–3.78]). Median age at diagnosis for sporadic (59 years) and familial (49 years) cases differed significantly (p < 0.001) with or without TP53 mutations. Age at diagnosis between the two types of familial carriers were not significantly different, with median age of 47 for 1675delA and 52.5 for 1135insA carriers (p = 0.245). For cases ≥50 years at diagnosis, a trend toward longer survival for sporadic over familial cases was observed (p = 0.08). The opposite trend was observed for cases Conclusion There do not seem to be a protective advantage for familial BRCA1 carriers without TP53 mutations over familial cases with TP53 mutations. However, there seem to be a trend towards initial advantage in survival for familial cases compared to sporadic cases diagnosed before the age of 50 both with and without TP53 mutations. However, this trend diminishes over time and for cases diagnosed ≥50 years the sporadic cases show a trend towards an advantage in survival over familial cases. Although this data set is small, if confirmed, this may be a link in the evidence that the differences in ovarian cancer survival reported, are not due to the type of BRCA1 mutation, but may be secondary to genetic factors shared. This may have clinical implications for follow-up such as prophylactic surgery within carriers of the two most frequent Norwegian BRCA1 founder mutations.

Published
2005
Full Text
View/download PDF

11. Effect of the codon 72 polymorphism (c.215G>C, p.Arg72Pro) in combination with somatic sequence variants in theTP53 gene on survival in patients with advanced ovarian carcinoma

Author

Wang, Yun, primary, Kringen, Pedro, additional, Kristensen, Gunnar B., additional, Holm, Ruth, additional, Baekelandt, Mark M.O., additional, Olivier, Magali, additional, Skomedal, Hanne, additional, Hainaut, Pierre, additional, Tropé, Claes G., additional, Abeler, Vera M., additional, Nesland, Jahn M., additional, Børresen-Dale, Anne-Lise, additional, and Helland, Åslaug, additional

Published
2006
Full Text
View/download PDF

14. Effect of the codon 72 polymorphism (c.215G>C, p.Arg72Pro) in combination with somatic sequence variants in theTP53gene on survival in patients with advanced ovarian carcinoma

Author

Wang, Yun, primary, Kringen, Pedro, additional, Kristensen, Gunnar B., additional, Holm, Ruth, additional, Baekelandt, Mark M.O., additional, Olivier, Magali, additional, Skomedal, Hanne, additional, Hainaut, Pierre, additional, Tropé, Claes G., additional, Abeler, Vera M., additional, Nesland, Jahn M., additional, Børresen-Dale, Anne-Lise, additional, and Helland, Åslaug, additional

Published
2004
Full Text
View/download PDF

16. Effect of the codon 72 polymorphism (c.215G>C, p.Arg72Pro) in combination with somatic sequence variants in the TP53 gene on survival in patients with advanced ovarian carcinoma.

Author

Wang, Yun, Kringen, Pedro, Kristensen, Gunnar B., Holm, Ruth, Baekelandt, Mark M.O., Olivier, Magali, Skomedal, Hanne, Hainaut, Pierre, Tropé, Claes G., Abeler, Vera M., Nesland, Jahn M., Børresen-Dale, Anne-Lise, and Helland, Åslaug

Abstract

This study was conducted to evaluate the frequency and prognostic impact of TP53 alterations stratified for the TP53 codon 72 polymorphism (c.215G>C, p.Arg72Pro) in a cohort of 109 patients with advanced ovarian carcinomas. TP53 sequence variants were observed in 80 of the 109 (73.4%) tumors and were significantly associated with grade of differentiation (P=0.001). A tendency towards higher frequency of sequence variants in tumors with higher FIGO stages was seen (P=0.05). The type of TP53 sequence variant (transition A:T>G:C vs. G:C>A:T at CpG dinucleotides, and transversion G:C>T:A) had significant correlation with patients' age (P=0.04) with more A:T>G:C in patients over 60 years old. No significant associations were found between frequency of sequence variants and age at diagnosis, histological type, size of residual tumor after primary surgery, or long-term survival. Analyses of the codon 72 polymorphism in tumor DNA gave a higher frequency of homozygosity/hemizygosity than expected from the population frequency, particularly for the Pro allele. Tumors homozygous or hemizygous for the Pro allele had significantly higher frequency of TP53 sequence variants, particularly of the nonmissense type (P=0.002), and patients with these types of alterations had significantly shorter survival (P=0.04). TP53 protein accumulation, determined by immunohistochemistry (IHC), was found in 67.9% (74 out of 109) of the tumors, was significantly more common among serous than nonserous ovarian carcinomas (P=0.008), and had a significant effect on progression-free survival (P=0.03). p63 (TP73L; formerly TP63) and p73 (TP73) protein accumulation detected by IHC was seen in 67.9 and 0% of the tumors, respectively. A significantly higher frequency of p63-positive cases was seen among serous tumors (P=0.008) and tended to increase with increasing FIGO stages (P=0.05), but had no significant effect on survival. No association between p63 protein accumulation and TP53 protein accumulation was seen. Hum Mutat 24:21-34, 2004. © 2004 Wiley-Liss, Inc. [ABSTRACT FROM AUTHOR]

Published
2004
Full Text
View/download PDF

17. Evaluation of arrayed primer extension for TP53mutation detection in breast and ovarian carcinomas

Author

Kringen, Pedro, Bergamaschi, Anna, Undlien Due, Eldri, Wang, Yun, Tagliabue, Elda, Nesland, Jahn M., Nehman, Aune, Tönisson, Neeme, and Børresen-Dale, Anne-Lise

Abstract

Mutations in the tumor suppressor gene TP53 are associated with a wide range of different cancers and may have prognostic and therapeutic implications. Methods for rapid and sensitive detection of mutations in this gene are therefore required. In order to make screening more effective, a commercially available TP53 genotyping microarray from Asper Biotech has been constructed by arrayed primer extension (APEX). The present study is the first report that blindly evaluates the efficiency of the second generation APEX TP53 genotype chip outside the Asper laboratory and compares it to temporal temperature gradient electrophoresis (TTGE) and sequencing of TP53 for mutation detection in ovarian and breast cancer samples. All nucleotides in the TP53 gene from exon 2–9 are included on the chip by synthesis and application of sequence-specific oligonucleotides. The chip was validated by screening 48 breast and 11 ovarian cancer cases, all of which had previously been analyzed by TTGE and sequencing. APEX scored 17 of 20 sequence variants, missing one deletion, one insertion, and a missense mutation. Resequencing efficiency using APEX was 92% for both DNA strands and 99.5% for sense and/or antisense strand. We conclude that the APEX TP53 microarray is a robust, rapid, and comprehensive screening tool for sequence alterations in tumors.

Published
2005
Full Text
View/download PDF

18. On Assessing Trustworthy AI in Healthcare. Machine Learning as a Supportive Tool to Recognize Cardiac Arrest in Emergency Calls

Author

Zicari, Roberto V, Brusseau, James, Blomberg, Stig Nikolaj, Christensen, Helle Collatz, Coffee, Megan, Ganapini, Marianna B, Gerke, Sara, Gilbert, Thomas Krendl, Hickman, Eleanore, Hildt, Elisabeth, Holm, Sune, Kühne, Ulrich, Madai, Vince I, Osika, Walter, Spezzatti, Andy, Schnebel, Eberhard, Tithi, Jesmin Jahan, Vetter, Dennis, Westerlund, Magnus, Wurth, Renee, Amann, Julia, Antun, Vegard, Beretta, Valentina, Bruneault, Frédérick, Campano, Erik, Düdder, Boris, Gallucci, Alessio, Goffi, Emmanuel, Haase, Christoffer Bjerre, Hagendorff, Thilo, Kringen, Pedro, Möslein, Florian, Ottenheimer, Davi, Ozols, Matiss, Palazzani, Laura, Petrin, Martin, Tafur, Karin, Tørresen, Jim, Volland, Holger, and Kararigas, Georgios

Subjects
ComputingMethodologies_PATTERNRECOGNITION, 46 Information and Computing Sciences, 8.3 Policy, ethics, and research governance, 42 Health Sciences, 3 Good Health and Well Being, 8 Health and social care services research, 4203 Health Services and Systems, Cardiovascular, GeneralLiterature_MISCELLANEOUS, 3. Good health
Abstract

Artificial Intelligence (AI) has the potential to greatly improve the delivery of healthcare and other services that advance population health and wellbeing. However, the use of AI in healthcare also brings potential risks that may cause unintended harm. To guide future developments in AI, the High-Level Expert Group on AI set up by the European Commission (EC), recently published ethics guidelines for what it terms “trustworthy” AI. These guidelines are aimed at a variety of stakeholders, especially guiding practitioners toward more ethical and more robust applications of AI. In line with efforts of the EC, AI ethics scholarship focuses increasingly on converting abstract principles into actionable recommendations. However, the interpretation, relevance, and implementation of trustworthy AI depend on the domain and the context in which the AI system is used. The main contribution of this paper is to demonstrate how to use the general AI HLEG trustworthy AI guidelines in practice in the healthcare domain. To this end, we present a best practice of assessing the use of machine learning as a supportive tool to recognize cardiac arrest in emergency calls. The AI system under assessment is currently in use in the city of Copenhagen in Denmark. The assessment is accomplished by an independent team composed of philosophers, policy makers, social scientists, technical, legal, and medical experts. By leveraging an interdisciplinary team, we aim to expose the complex trade-offs and the necessity for such thorough human review when tackling socio-technical applications of AI in healthcare. For the assessment, we use a process to assess trustworthy AI, called 1Z-Inspection® to identify specific challenges and potential ethical trade-offs when we consider AI in practice.

19. On Assessing Trustworthy AI in Healthcare: Machine Learning as a Supportive Tool to Recognize Cardiac Arrest in Emergency Calls

Author

Zicari, Roberto V., Brusseau, James, Blomberg, Stig N., Collatz Christensen, Helle, Coffee, Megan, Ganapini, Marianna B., Gerke, Sara, Krendl Gilbert, Thomas, Hickman, Eleanore, Hildt, Elisabeth, Holm, Sune, Kühne, Ulrich, Madai, Vince I., Osika, Walter, Spezzatti, Andy, Schnebel, Eberhard, Tithi, Jesmin J., Vetter, Dennis, Westerlund, Magnus, Wurth, Renee, Amann, Julia, Vegard, Antun, Beretta, Valentina, Bruneault, Frédérick, Campano, Erik, Düdder, Boris, Gallucci, Alessio, Goffi, Emmanuel, Haase, Christoffer B., Hagendorff, Thilo, Kringen, Pedro, Möslein, Florian, Ottenheimer, Davi, Ozols, Matiss, Palazzani, Laura, Petrin, Martin, Tafur, Karin, Tørresen, Jim, Volland, Holger, and Kararigas, Georgios

Subjects
Artificial intelligence, ComputingMethodologies_PATTERNRECOGNITION, Ethical trade-off, Case study, Explainable AI, Healthcare, Cardiac arrest, Trust, GeneralLiterature_MISCELLANEOUS, Trustworthy AI, 3. Good health
Abstract

Artificial Intelligence (AI) has the potential to greatly improve the delivery of healthcare and other services that advance population health and wellbeing. However, the use of AI in healthcare also brings potential risks that may cause unintended harm. To guide future developments in AI, the High-Level Expert Group on AI set up by the European Commission (EC), recently published ethics guidelines for what it terms “trustworthy” AI. These guidelines are aimed at a variety of stakeholders, especially guiding practitioners toward more ethical and more robust applications of AI. In line with efforts of the EC, AI ethics scholarship focuses increasingly on converting abstract principles into actionable recommendations. However, the interpretation, relevance, and implementation of trustworthy AI depend on the domain and the context in which the AI system is used. The main contribution of this paper is to demonstrate how to use the general AI HLEG trustworthy AI guidelines in practice in the healthcare domain. To this end, we present a best practice of assessing the use of machine learning as a supportive tool to recognize cardiac arrest in emergency calls. The AI system under assessment is currently in use in the city of Copenhagen in Denmark. The assessment is accomplished by an independent team composed of philosophers, policy makers, social scientists, technical, legal, and medical experts. By leveraging an interdisciplinary team, we aim to expose the complex trade-offs and the necessity for such thorough human review when tackling socio-technical applications of AI in healthcare. For the assessment, we use a process to assess trustworthy AI, called 1Z-Inspection® to identify specific challenges and potential ethical trade-offs when we consider AI in practice., Frontiers in Human Dynamics, 3, ISSN:2673-2726

Catalog

Books, media, physical & digital resources
See catalog results