Jurriaan Barkey Wolf
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Jurriaan Barkey Wolf has a total of 66 co-authors. Here are the top co-authors by number of shared publications:
- Nanda Horeweg
Department of Radiation Oncology, Leiden University Medical Center, Leiden, The Netherlands
5 shared publications - Tjalling Bosse
Department of Pathology, Leiden University Medical Center, Leiden, The Netherlands
5 shared publications - Viktor Hendrik Koelzer
University Hospital Basel
5 shared publications - Marco de Bruyn
Department of Obstetrics and Gynaecology, University Medical Center Groningen, University of Groningen, Netherlands
4 shared publications - Sarah Volinsky-Fremond
Department of Pathology, Leiden University Medical Center, Leiden, The Netherlands
4 shared publications Cor D. de Kroon
Department of Obstetrics and Gynaecology, Leiden University Medical Center, Leiden, the Netherlands
3 shared publications
Latest Publications
- Open Access
Leapfrogging cancer diagnostics with AI: a mixed-methods assessment of implementation determinants in low- and middle-income countries
Introduction: Cancer incidence is rising rapidly in low- and middle-income countries (LMICs), where diagnostic pathology capacity is constrained. AI applied to routine histopathology could strengthen diagnostic capacity and offer leapfrogging potential by embedding AI into diagnostic systems as they are built up. However, without empirically identifying implementation determinants in LMIC cancer-diagnostic settings, AI solutions risk being poorly aligned with local constraints, limiting impact or exacerbating existing disparities. Methods: We conducted an exploratory mixed-methods pilot study across three phases: 1) a structured survey characterising the diagnostic and technical landscape and AI priorities; 2) focus group discussions (FGDs) to identify implementation determinants and 3) a second survey quantifying FGD-derived themes. Twelve physicians (nine pathologists, three gynaecologists) were purposively recruited (10 LMIC, 2 high-income countries for reference) and findings integrated through narrative triangulation. Results: Diagnostic and AI infrastructure varied substantially and did not track consistently with country income classification. H&E staining was universally available, immunohistochemistry capacity was below WHO-recommended capacity for cancer diagnosis at 7 of 10 sites and molecular testing absent or minimal at 6 of 10 sites. Whole-slide-imaging scanners were available at 8 of 10 sites, graphics processing units, required for efficient AI inference, were available at only 4 of 10 sites. Participants prioritised AI for cancer screening, diagnosis, and biomarker prediction from H&E slides. Six implementation determinants were identified: AI literacy, governance frameworks, funding, evidence generation, clinical deployment, and collaboration. Financial constraints emerged as the dominant barrier across all determinants, and collaboration as a key enabler. Conclusion: AI implementation readiness varied substantially within income groups, suggesting that country income classification is an insufficient proxy for institutional readiness, underscoring the need for institution-level assessment. Realising the leapfrogging potential of AI in this context requires coordinated progress across all determinants, with sustainable financing, and collaboration as enablers of the other determinants. Key messages: What is already known on this topic?: Cancer incidence is rising rapidly in LMICs, where diagnostic capacity and molecular testing capabilities are severely constrained. AI could alleviate gaps, yet evidence on implementation determinants for integrating AI into cancer diagnostics in these settings remains limited. What this study adds?: In this exploratory sample, AI implementation readiness varied substantially across LMIC settings, including within the same country income group. Readiness is shaped by six implementation determinants: AI literacy, governance frameworks, funding, evidence generation, clinical deployment, and collaboration, with financial constraints as the dominant cross-cutting barrier. How this study might affect research, practice or policy?: These findings support institution-level feasibility assessment rather than reliance on country income classification as a proxy and highlight the need for sustainable financing and collaboration as foundational prerequisites for AI implementation in cancer diagnostics in LMICs.
Open AccessLeapfrogging cancer diagnostics with AI: a mixed-methods assessment of implementation determinants in low- and middle-income countries
Introduction: Cancer incidence is rising rapidly in low- and middle-income countries (LMICs), where diagnostic pathology capacity is constrained. AI applied to routine histopathology could strengthen diagnostic capacity and offer leapfrogging potential by embedding AI into diagnostic systems as they are built up. However, without empirically identifying implementation determinants in LMIC cancer-diagnostic settings, AI solutions risk being poorly aligned with local constraints, limiting impact or exacerbating existing disparities. Methods: We conducted an exploratory mixed-methods pilot study across three phases: 1) a structured survey characterising the diagnostic and technical landscape and AI priorities; 2) focus group discussions (FGDs) to identify implementation determinants and 3) a second survey quantifying FGD-derived themes. Twelve physicians (nine pathologists, three gynaecologists) were purposively recruited (10 LMIC, 2 high-income countries for reference) and findings integrated through narrative triangulation. Results: Diagnostic and AI infrastructure varied substantially and did not track consistently with country income classification. H&E staining was universally available, immunohistochemistry capacity was below WHO-recommended capacity for cancer diagnosis at 7 of 10 sites and molecular testing absent or minimal at 6 of 10 sites. Whole-slide-imaging scanners were available at 8 of 10 sites, graphics processing units, required for efficient AI inference, were available at only 4 of 10 sites. Participants prioritised AI for cancer screening, diagnosis, and biomarker prediction from H&E slides. Six implementation determinants were identified: AI literacy, governance frameworks, funding, evidence generation, clinical deployment, and collaboration. Financial constraints emerged as the dominant barrier across all determinants, and collaboration as a key enabler. Conclusion: AI implementation readiness varied substantially within income groups, suggesting that country income classification is an insufficient proxy for institutional readiness, underscoring the need for institution-level assessment. Realising the leapfrogging potential of AI in this context requires coordinated progress across all determinants, with sustainable financing, and collaboration as enablers of the other determinants. Key messages: What is already known on this topic?: Cancer incidence is rising rapidly in LMICs, where diagnostic capacity and molecular testing capabilities are severely constrained. AI could alleviate gaps, yet evidence on implementation determinants for integrating AI into cancer diagnostics in these settings remains limited. What this study adds?: In this exploratory sample, AI implementation readiness varied substantially across LMIC settings, including within the same country income group. Readiness is shaped by six implementation determinants: AI literacy, governance frameworks, funding, evidence generation, clinical deployment, and collaboration, with financial constraints as the dominant cross-cutting barrier. How this study might affect research, practice or policy?: These findings support institution-level feasibility assessment rather than reliance on country income classification as a proxy and highlight the need for sustainable financing and collaboration as foundational prerequisites for AI implementation in cancer diagnostics in LMICs.
- Open Access
Deep Learning-Based Screening for POLE mutations on Histopathology Slides in Endometrial Cancer
POLE sequencing for somatic mutations ( POLE mut) guides adjuvant therapy in endometrial cancer (EC), but cost and infrastructural considerations lead to limited uptake. Omission of POLE testing leads to unnecessary exposure to radiotherapy and/or chemotherapy. We developed POLARIX, a multiple instance deep learning model with attention pooling, which predicts POLE mutation status from routine hematoxylin and eosin whole-slide images (WSIs). Trained on 2,238 cases from eleven EC cohorts, POLARIX showed clinical-grade discrimination across three external cohorts (Pooled: AUC=0.95, 95% CI: 0.91–0.98; n =68/481 POLE mut/ POLE wt). Attention maps highlight POLE morphologies. Clinical applicability is demonstrated using predefined thresholds based on three resource scenarios. The most sensitive threshold (“Low”) yields a test reduction of 77% (73%-81%) (sensitivity: 93% (85%-99%), specificity: 89% (87%-92%)). POLARIX is an interpretable and cost-efficient approach to reduce POLE testing in women with endometrial cancer, broadening access to precision oncology.
Open AccessDeep Learning-Based Screening for POLE mutations on Histopathology Slides in Endometrial Cancer
POLE sequencing for somatic mutations ( POLE mut) guides adjuvant therapy in endometrial cancer (EC), but cost and infrastructural considerations lead to limited uptake. Omission of POLE testing leads to unnecessary exposure to radiotherapy and/or chemotherapy. We developed POLARIX, a multiple instance deep learning model with attention pooling, which predicts POLE mutation status from routine hematoxylin and eosin whole-slide images (WSIs). Trained on 2,238 cases from eleven EC cohorts, POLARIX showed clinical-grade discrimination across three external cohorts (Pooled: AUC=0.95, 95% CI: 0.91–0.98; n =68/481 POLE mut/ POLE wt). Attention maps highlight POLE morphologies. Clinical applicability is demonstrated using predefined thresholds based on three resource scenarios. The most sensitive threshold (“Low”) yields a test reduction of 77% (73%-81%) (sensitivity: 93% (85%-99%), specificity: 89% (87%-92%)). POLARIX is an interpretable and cost-efficient approach to reduce POLE testing in women with endometrial cancer, broadening access to precision oncology.
- Open Access
- 1 July 2024
- Published by Springer Nature in Nature Medicine
- Vol. 30 (7) , 2092
Author Correction: Prediction of recurrence risk in endometrial cancer with multimodal deep learning
- 1 July 2024
- Published by Springer Nature in Nature Medicine
- Vol. 30 (7) , 2092
Open AccessAuthor Correction: Prediction of recurrence risk in endometrial cancer with multimodal deep learning
- Top Cited PapersOpen Access
- 24 May 2024
- Published by Springer Nature in Nature Medicine
- Vol. 30 (7) , 1962-1973
Prediction of recurrence risk in endometrial cancer with multimodal deep learning
Predicting distant recurrence of endometrial cancer (EC) is crucial for personalized adjuvant treatment. The current gold standard of combined pathological and molecular profiling is costly, hampering implementation. Here we developed HECTOR (histopathology-based endometrial cancer tailored outcome risk), a multimodal deep learning prognostic model using hematoxylin and eosin-stained, whole-slide images and tumor stage as input, on 2,072 patients from eight EC cohorts including the PORTEC-1/-2/-3 randomized trials. HECTOR demonstrated C-indices in internal (n = 353) and two external (n = 160 andn = 151) test sets of 0.789, 0.828 and 0.815, respectively, outperforming the current gold standard, and identified patients with markedly different outcomes (10-year distant recurrence-free probabilities of 97.0%, 77.7% and 58.1% for HECTOR low-, intermediate- and high-risk groups, respectively, by Kaplan–Meier analysis). HECTOR also predicted adjuvant chemotherapy benefit better than current methods. Morphological and genomic feature extraction identified correlates of HECTOR risk groups, some with therapeutic potential. HECTOR improves on the current gold standard and may help delivery of personalized treatment in EC.
- 24 May 2024
- Published by Springer Nature in Nature Medicine
- Vol. 30 (7) , 1962-1973
Top Cited PapersOpen AccessPrediction of recurrence risk in endometrial cancer with multimodal deep learning
Predicting distant recurrence of endometrial cancer (EC) is crucial for personalized adjuvant treatment. The current gold standard of combined pathological and molecular profiling is costly, hampering implementation. Here we developed HECTOR (histopathology-based endometrial cancer tailored outcome risk), a multimodal deep learning prognostic model using hematoxylin and eosin-stained, whole-slide images and tumor stage as input, on 2,072 patients from eight EC cohorts including the PORTEC-1/-2/-3 randomized trials. HECTOR demonstrated C-indices in internal (n = 353) and two external (n = 160 andn = 151) test sets of 0.789, 0.828 and 0.815, respectively, outperforming the current gold standard, and identified patients with markedly different outcomes (10-year distant recurrence-free probabilities of 97.0%, 77.7% and 58.1% for HECTOR low-, intermediate- and high-risk groups, respectively, by Kaplan–Meier analysis). HECTOR also predicted adjuvant chemotherapy benefit better than current methods. Morphological and genomic feature extraction identified correlates of HECTOR risk groups, some with therapeutic potential. HECTOR improves on the current gold standard and may help delivery of personalized treatment in EC.
- Open Access
HECTOR: multimodal deep learning predicts recurrence risk in endometrial cancer
Predicting distant recurrence of endometrial cancer (EC) is crucial for personalized adjuvant treatment. The current gold standard of combined pathological and molecular profiling is costly, hampering implementation. We developed HECTOR ( H istopathology-based E ndometrial C ancer T ailored O utcome R isk), a multimodal deep learning prognostic model using hematoxylin-and-eosin-stained whole-slide-images and tumor stage as input, on 1,912 patients from seven EC cohorts including the PORTEC-1/-2/-3 randomized trials. HECTOR demonstrated C-indices in internal ( n = 353) and external ( n = 151) test sets of 0.788 and 0.816 respectively, outperforming the current gold-standard, and identified patients with markedly different outcomes (10-year distant recurrence-free probabilities of 97.0%, 77.7% and 58.1% for HECTOR low, intermediate and high risk groups). HECTOR also predicted adjuvant chemotherapy benefit better than current methods. Morphological and genomic feature extraction identified correlates of HECTOR risk groups, some with therapeutic potential. HECTOR improves on the current gold-standard and may help delivery of personalized treatment in EC.
Open AccessHECTOR: multimodal deep learning predicts recurrence risk in endometrial cancer
Predicting distant recurrence of endometrial cancer (EC) is crucial for personalized adjuvant treatment. The current gold standard of combined pathological and molecular profiling is costly, hampering implementation. We developed HECTOR ( H istopathology-based E ndometrial C ancer T ailored O utcome R isk), a multimodal deep learning prognostic model using hematoxylin-and-eosin-stained whole-slide-images and tumor stage as input, on 1,912 patients from seven EC cohorts including the PORTEC-1/-2/-3 randomized trials. HECTOR demonstrated C-indices in internal ( n = 353) and external ( n = 151) test sets of 0.788 and 0.816 respectively, outperforming the current gold-standard, and identified patients with markedly different outcomes (10-year distant recurrence-free probabilities of 97.0%, 77.7% and 58.1% for HECTOR low, intermediate and high risk groups). HECTOR also predicted adjuvant chemotherapy benefit better than current methods. Morphological and genomic feature extraction identified correlates of HECTOR risk groups, some with therapeutic potential. HECTOR improves on the current gold-standard and may help delivery of personalized treatment in EC.