-
Je něco špatně v tomto záznamu ?
International multicenter validation of AI-driven ultrasound detection of ovarian cancer
F. Christiansen, E. Konuk, AR. Ganeshan, R. Welch, J. Palés Huix, A. Czekierdowski, FPG. Leone, LA. Haak, R. Fruscio, A. Gaurilcikas, D. Franchi, D. Fischerova, E. Mor, L. Savelli, MÀ. Pascual, MJ. Kudla, S. Guerriero, F. Buonomo, K. Liuba, N....
Jazyk angličtina Země Spojené státy americké
Typ dokumentu časopisecké články, multicentrická studie, validační studie
Grantová podpora
231143
Radiumhemmets Forskningsfonder (Cancer Research Foundations of Radiumhemmet)
211657 Pi 01 H
Cancerfonden (Swedish Cancer Society)
2020-01702
Vetenskapsrdet (Swedish Research Council)
- MeSH
- deep learning MeSH
- dospělí MeSH
- lidé středního věku MeSH
- lidé MeSH
- nádory vaječníků * diagnostické zobrazování MeSH
- neuronové sítě * MeSH
- retrospektivní studie MeSH
- senioři MeSH
- senzitivita a specificita MeSH
- ultrasonografie * metody MeSH
- umělá inteligence MeSH
- Check Tag
- dospělí MeSH
- lidé středního věku MeSH
- lidé MeSH
- senioři MeSH
- ženské pohlaví MeSH
- Publikační typ
- časopisecké články MeSH
- multicentrická studie MeSH
- validační studie MeSH
Ovarian lesions are common and often incidentally detected. A critical shortage of expert ultrasound examiners has raised concerns of unnecessary interventions and delayed cancer diagnoses. Deep learning has shown promising results in the detection of ovarian cancer in ultrasound images; however, external validation is lacking. In this international multicenter retrospective study, we developed and validated transformer-based neural network models using a comprehensive dataset of 17,119 ultrasound images from 3,652 patients across 20 centers in eight countries. Using a leave-one-center-out cross-validation scheme, for each center in turn, we trained a model using data from the remaining centers. The models demonstrated robust performance across centers, ultrasound systems, histological diagnoses and patient age groups, significantly outperforming both expert and non-expert examiners on all evaluated metrics, namely F1 score, sensitivity, specificity, accuracy, Cohen's kappa, Matthew's correlation coefficient, diagnostic odds ratio and Youden's J statistic. Furthermore, in a retrospective triage simulation, artificial intelligence (AI)-driven diagnostic support reduced referrals to experts by 63% while significantly surpassing the diagnostic performance of the current practice. These results show that transformer-based models exhibit strong generalization and above human expert-level diagnostic accuracy, with the potential to alleviate the shortage of expert ultrasound examiners and improve patient outcomes.
3rd Faculty of Medicine Charles University Prague Czech Republic
Department of Clinical Science and Education Södersjukhuset Karolinska Institutet Stockholm Sweden
Department of Gynecological Oncology and Gynecology Medical University of Lublin Lublin Poland
Department of Medicine and Surgery University of Milan Bicocca Milan Italy
Department of Obstetrics and Gynaecology Lithuanian University of Health Sciences Kaunas Lithuania
Department of Obstetrics and Gynecology Clínica Universidad de Navarra Pamplona Spain
Department of Obstetrics and Gynecology Rizal Medical Center Manila Philippines
Department of Obstetrics and Gynecology Skåne University Hospital Lund Sweden
Department of Obstetrics and Gynecology Södersjukhuset Stockholm Sweden
Department of Obstetrics Gynecology and Reproduction Dexeus University Hospital Barcelona Spain
Digital Futures KTH Royal Institute of Technology Stockholm Sweden
Fondazione Poliambulanza Istituto Ospedaliero Brescia Italy
Gynecologic and Obstetric Unit Women's and Children's Department Forlì Hospital Forlì Italy
Gynecology and Breast Care Center Mater Olbia Hospital Olbia Italy
Institute for Maternal and Child Health IRCCS 'Burlo Garofolo' Trieste Italy
Institute for the Care of Mother and Child Prague Czech Republic
Obstetrics and Gynecology Unit Forlì and Faenza Hospitals AUSL Romagna Forlì Italy
Science for Life Laboratory Stockholm Sweden
Unit of Preventive Gynecology European Institute of Oncology IRCCS Milan Italy
UO Gynecology Fondazione IRCCS San Gerardo dei Tintori Monza Italy
Citace poskytuje Crossref.org
- 000
- 00000naa a2200000 a 4500
- 001
- bmc25010418
- 003
- CZ-PrNML
- 005
- 20250429134746.0
- 007
- ta
- 008
- 250415s2025 xxu f 000 0|eng||
- 009
- AR
- 024 7_
- $a 10.1038/s41591-024-03329-4 $2 doi
- 035 __
- $a (PubMed)39747679
- 040 __
- $a ABA008 $b cze $d ABA008 $e AACR2
- 041 0_
- $a eng
- 044 __
- $a xxu
- 100 1_
- $a Christiansen, Filip $u Department of Clinical Science and Education, Södersjukhuset, Karolinska Institutet, Stockholm, Sweden $u Department of Obstetrics and Gynecology, Södersjukhuset, Stockholm, Sweden $u School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Stockholm, Sweden $u Science for Life Laboratory, Stockholm, Sweden $1 https://orcid.org/0000000172069611
- 245 10
- $a International multicenter validation of AI-driven ultrasound detection of ovarian cancer / $c F. Christiansen, E. Konuk, AR. Ganeshan, R. Welch, J. Palés Huix, A. Czekierdowski, FPG. Leone, LA. Haak, R. Fruscio, A. Gaurilcikas, D. Franchi, D. Fischerova, E. Mor, L. Savelli, MÀ. Pascual, MJ. Kudla, S. Guerriero, F. Buonomo, K. Liuba, N. Montik, JL. Alcázar, E. Domali, NCP. Pangilinan, C. Carella, M. Munaretto, P. Saskova, D. Verri, C. Visenzi, P. Herman, K. Smith, E. Epstein
- 520 9_
- $a Ovarian lesions are common and often incidentally detected. A critical shortage of expert ultrasound examiners has raised concerns of unnecessary interventions and delayed cancer diagnoses. Deep learning has shown promising results in the detection of ovarian cancer in ultrasound images; however, external validation is lacking. In this international multicenter retrospective study, we developed and validated transformer-based neural network models using a comprehensive dataset of 17,119 ultrasound images from 3,652 patients across 20 centers in eight countries. Using a leave-one-center-out cross-validation scheme, for each center in turn, we trained a model using data from the remaining centers. The models demonstrated robust performance across centers, ultrasound systems, histological diagnoses and patient age groups, significantly outperforming both expert and non-expert examiners on all evaluated metrics, namely F1 score, sensitivity, specificity, accuracy, Cohen's kappa, Matthew's correlation coefficient, diagnostic odds ratio and Youden's J statistic. Furthermore, in a retrospective triage simulation, artificial intelligence (AI)-driven diagnostic support reduced referrals to experts by 63% while significantly surpassing the diagnostic performance of the current practice. These results show that transformer-based models exhibit strong generalization and above human expert-level diagnostic accuracy, with the potential to alleviate the shortage of expert ultrasound examiners and improve patient outcomes.
- 650 _2
- $a lidé $7 D006801
- 650 _2
- $a ženské pohlaví $7 D005260
- 650 12
- $a nádory vaječníků $x diagnostické zobrazování $7 D010051
- 650 _2
- $a retrospektivní studie $7 D012189
- 650 12
- $a ultrasonografie $x metody $7 D014463
- 650 _2
- $a lidé středního věku $7 D008875
- 650 12
- $a neuronové sítě $7 D016571
- 650 _2
- $a dospělí $7 D000328
- 650 _2
- $a umělá inteligence $7 D001185
- 650 _2
- $a senioři $7 D000368
- 650 _2
- $a deep learning $7 D000077321
- 650 _2
- $a senzitivita a specificita $7 D012680
- 655 _2
- $a časopisecké články $7 D016428
- 655 _2
- $a multicentrická studie $7 D016448
- 655 _2
- $a validační studie $7 D023361
- 700 1_
- $a Konuk, Emir $u School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Stockholm, Sweden $u Science for Life Laboratory, Stockholm, Sweden $1 https://orcid.org/0000000194374553
- 700 1_
- $a Ganeshan, Adithya Raju $u Department of Clinical Science and Education, Södersjukhuset, Karolinska Institutet, Stockholm, Sweden $u School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Stockholm, Sweden $u Science for Life Laboratory, Stockholm, Sweden $1 https://orcid.org/0000000182166458
- 700 1_
- $a Welch, Robert $u Department of Clinical Science and Education, Södersjukhuset, Karolinska Institutet, Stockholm, Sweden $u School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Stockholm, Sweden $u Science for Life Laboratory, Stockholm, Sweden $1 https://orcid.org/0000000318196120
- 700 1_
- $a Palés Huix, Joana $u School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Stockholm, Sweden $u Science for Life Laboratory, Stockholm, Sweden $1 https://orcid.org/0009000841171638
- 700 1_
- $a Czekierdowski, Artur $u Department of Gynecological Oncology and Gynecology, Medical University of Lublin, Lublin, Poland
- 700 1_
- $a Leone, Francesco Paolo Giuseppe $u Unit of Obstetrics & Gynecology, Department of Biomedical and Clinical Sciences, Luigi Sacco University Hospital, University of Milan, Milan, Italy
- 700 1_
- $a Haak, Lucia Anna $u Institute for the Care of Mother and Child, Prague, Czech Republic $u Third Faculty of Medicine, Charles University, Prague, Czech Republic $1 https://orcid.org/0000000157497968
- 700 1_
- $a Fruscio, Robert $u Department of Medicine and Surgery, University of Milan-Bicocca, Milan, Italy $u UO Gynecology, Fondazione IRCCS San Gerardo dei Tintori, Monza, Italy $1 https://orcid.org/0000000156882194
- 700 1_
- $a Gaurilcikas, Adrius $u Department of Obstetrics and Gynaecology, Lithuanian University of Health Sciences, Kaunas, Lithuania
- 700 1_
- $a Franchi, Dorella $u Unit of Preventive Gynecology, European Institute of Oncology IRCCS, Milan, Italy $1 https://orcid.org/0000000239505538
- 700 1_
- $a Fischerova, Daniela $u Gynecologic Oncology Centre, Department of Gynecology, Obstetrics and Neonatology, First Faculty of Medicine, Charles University and General University Hospital in Prague, Prague, Czech Republic $1 https://orcid.org/0000000272243218 $7 xx0074804
- 700 1_
- $a Mor, Elisa $u Fondazione Poliambulanza Istituto Ospedaliero, Brescia, Italy
- 700 1_
- $a Savelli, Luca $u Obstetrics and Gynecology Unit, Forlì and Faenza Hospitals, AUSL Romagna, Forlì, Italy
- 700 1_
- $a Pascual, Maria Àngela $u Department of Obstetrics, Gynecology, and Reproduction, Dexeus University Hospital, Barcelona, Spain $1 https://orcid.org/0000000150956981
- 700 1_
- $a Kudla, Marek Jerzy $u Department of Perinatology and Oncological Gynecology, Faculty of Medical Sciences, Medical University of Silesia, Katowice, Poland
- 700 1_
- $a Guerriero, Stefano $u Centro Integrato di Procreazione Medicalmente Assistita e Diagnostica Ostetrico-Ginecologica, Azienda Ospedaliero Universitaria-Policlinico Duilio Casula, Monserrato, University of Cagliari, Cagliari, Italy
- 700 1_
- $a Buonomo, Francesca $u Institute for Maternal and Child Health, IRCCS 'Burlo Garofolo', Trieste, Italy $1 https://orcid.org/0000000265872622
- 700 1_
- $a Liuba, Karina $u Department of Obstetrics and Gynecology, Skåne University Hospital, Lund, Sweden
- 700 1_
- $a Montik, Nina $u Section of Obstetrics and Gynecology, Department of Clinical Sciences, Università Politecnica delle Marche, Azienda Ospedaliero-Universitaria delle Marche, Ancona, Italy
- 700 1_
- $a Alcázar, Juan Luis $u Department of Obstetrics and Gynecology, Clínica Universidad de Navarra, Pamplona, Spain
- 700 1_
- $a Domali, Ekaterini $u First Department of Obstetrics and Gynecology, Alexandra Hospital, Medical School, National and Kapodistrian University of Athens, Athens, Greece $1 https://orcid.org/0000000188993040
- 700 1_
- $a Pangilinan, Nelinda Catherine P $u Department of Obstetrics and Gynecology, Rizal Medical Center, Manila, Philippines
- 700 1_
- $a Carella, Chiara $u Unit of Obstetrics & Gynecology, Department of Biomedical and Clinical Sciences, Luigi Sacco University Hospital, University of Milan, Milan, Italy $1 https://orcid.org/0009000626905686
- 700 1_
- $a Munaretto, Maria $u Gynecologic and Obstetric Unit, Women's and Children's Department, Forlì Hospital, Forlì, Italy
- 700 1_
- $a Saskova, Petra $u Gynecologic Oncology Centre, Department of Gynecology, Obstetrics and Neonatology, First Faculty of Medicine, Charles University and General University Hospital in Prague, Prague, Czech Republic $1 https://orcid.org/0000000253678056
- 700 1_
- $a Verri, Debora $u Gynecology and Breast Care Center, Mater Olbia Hospital, Olbia, Italy $1 https://orcid.org/0000000274068804
- 700 1_
- $a Visenzi, Chiara $u Fondazione Poliambulanza Istituto Ospedaliero, Brescia, Italy
- 700 1_
- $a Herman, Pawel $u School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Stockholm, Sweden $u Digital Futures, KTH Royal Institute of Technology, Stockholm, Sweden
- 700 1_
- $a Smith, Kevin $u School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Stockholm, Sweden $u Science for Life Laboratory, Stockholm, Sweden
- 700 1_
- $a Epstein, Elisabeth $u Department of Clinical Science and Education, Södersjukhuset, Karolinska Institutet, Stockholm, Sweden. elisabeth.epstein@ki.se $u Department of Obstetrics and Gynecology, Södersjukhuset, Stockholm, Sweden. elisabeth.epstein@ki.se $1 https://orcid.org/0000000322987785
- 773 0_
- $w MED00003459 $t Nature medicine $x 1546-170X $g Roč. 31, č. 1 (2025), s. 189-196
- 856 41
- $u https://pubmed.ncbi.nlm.nih.gov/39747679 $y Pubmed
- 910 __
- $a ABA008 $b sig $c sign $y - $z 0
- 990 __
- $a 20250415 $b ABA008
- 991 __
- $a 20250429134742 $b ABA008
- 999 __
- $a ok $b bmc $g 2311647 $s 1247499
- BAS __
- $a 3
- BAS __
- $a PreBMC-MEDLINE
- BMC __
- $a 2025 $b 31 $c 1 $d 189-196 $e 20250102 $i 1546-170X $m Nature medicine $n Nat Med $x MED00003459
- GRA __
- $a 231143 $p Radiumhemmets Forskningsfonder (Cancer Research Foundations of Radiumhemmet)
- GRA __
- $a 211657 Pi 01 H $p Cancerfonden (Swedish Cancer Society)
- GRA __
- $a 2020-01702 $p Vetenskapsrdet (Swedish Research Council)
- LZP __
- $a Pubmed-20250415
