Meyke Hermsen

Meyke Hermsen

Auto-generated profile, based on publicly available information

Meyke Hermsen

Department of Pathology, Radboud University Medical Center, Nijmegen, The Netherlands

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 December 2017 in JAMA
Verified
Journal Article
Open access
PDF file
BBBabak Ehteshami Bejnordi
Babak Ehteshami Bejnordi
MVMitko Veta
Mitko Veta
PDPaul Johannes van Diest
Paul Johannes van Diest
BGBram van Ginneken
Bram van Ginneken
NKNico Karssemeijer
Nico Karssemeijer
RVRui Venâncio
Rui Venâncio
Quiz Ref ID Full digitalization of the microscopic evaluation of stained tissue sections in histopathology has become feasible in recent years because of advances in slide scanning technology and cost reduction in digital storage. Advantages of digital pathology include remote diagnostics, immediate availability of archival cases, and easier consultations with expert pathologists.1 Also, the possibility for computer-aided diagnostics may be advantageous.2

23 May 2016 in Scientific Reports
Verified
Journal Article
Open access
GLGeert Litjens
Geert Litjens
CSClara I. Sánchez
Clara I. Sánchez
NTNadya Timofeeva
Nadya Timofeeva
MHMeyke Hermsen
Meyke Hermsen
INIris D. Nagtegaal
Iris D. Nagtegaal
JLJeroen van der Laak
Jeroen van der Laak
Pathologists face a substantial increase in workload and complexity of histopathologic cancer diagnosis due to the advent of personalized medicine. Therefore, diagnostic protocols have to focus equally on efficiency and accuracy. In this paper we introduce ‘deep learning’ as a technique to improve the objectivity and efficiency of histopathologic slide analysis. Through two examples, prostate cancer identification in biopsy specimens and breast cancer metastasis detection in sentinel lymph nodes, we show the potential of this new methodology to reduce the workload for pathologists, while at the same time increasing objectivity of diagnoses. We found that all slides containing prostate cancer and micro- and macro-metastases of breast cancer could be identified automatically while 30–40% of the slides containing benign and normal tissue could be excluded without the use of any additional immunohistochemical markers or human intervention. We conclude that ‘deep learning’ holds great promise to improve the efficacy of prostate cancer diagnosis and breast cancer staging.

27 August 2018 in IEEE Transactions on Medical Imaging
Verified
Journal Article
Closed access
PBPeter Bandi
Peter Bandi
OGOscar Geessink
Oscar Geessink
QMQuirine Manson
Quirine Manson
MDMarcory Van Dijk
Marcory Van Dijk
MBMaschenka Balkenhol
Maschenka Balkenhol
GLGeert Litjens
Geert Litjens
Automated detection of cancer metastases in lymph nodes has the potential to improve assessment of prognosis for patients. To enable fair comparison between the algorithms for this purpose, we set up the CAMELYON17 challenge in conjunction with the IEEE International Symposium on Biomedical Imaging 2017 conference in Melbourne. Over 300 participants registered on the challenge website, of which 23 teams submitted a total of 37 algorithms before the initial deadline. Participants were provided with 899 whole-slide images for developing their algorithms. The developed algorithms were evaluated based on the test set encompassing 100 patients and 500 whole-slide images. The evaluation metric used was a quadratic weighted Cohen’s kappa. We discuss the algorithmic details of the ten best pre-conference and two post-conference submissions. All these participants used convolutional neural networks in combination with pre-and postprocessing steps. Algorithms differed mostly in neural network architecture, training strategy and pre-and postprocessing methodology. Overall, the kappa metric ranged from 0.89 to -0.13 across all submissions. The best results were obtained with pre-trained architectures such as ResNet. Confusion matrix analysis revealed that all participants struggled with reliably identifying isolated tumor cells, the smallest type of metastasis, with detection rates below 40%. Qualitative inspection of the results of the top participants showed categories of false positives, such as nerves or contamination, which could be targets for further optimization. Last, we show that simple combinations of the top algorithms result in higher kappa metric values than any algorithm individually, with 0.93 for the best combination.

5 September 2019 in Journal of the American Society of Nephrology
Verified
Journal Article
Closed access
PDF file
MHMeyke Hermsen
Meyke Hermsen
TBThomas de Bel
Thomas de Bel
MBMarjolijn Den Boer
Marjolijn Den Boer
ESEric J. Steenbergen
Eric J. Steenbergen
JKJesper Kers
Jesper Kers
Jeroen van der LaakJeroen van der Laak
Jeroen van der Laak
Significance Statement: Histopathologic assessment of kidney tissue currently relies on manual scoring or traditional image-processing techniques to quantify and classify tissue features, time-consuming approaches that have limited reproducibility. The authors present an alternative approach, featuring a convolutional neural network for multiclass segmentation of kidney tissue in sections stained by periodic acid–Schiff. Their findings demonstrate applicability of convolutional neural networks for tissue from multiple centers, for biopsies and nephrectomy samples, and for the analysis of both healthy and pathologic tissues. In addition, they validated the network’s results with components from the Banff classification system. Their convolutional neural network may have utility for quantitative studies involving kidney histopathology across centers and potential for application in routine diagnostics. Background: The development of deep neural networks is facilitating more advanced digital analysis of histopathologic images. We trained a convolutional neural network for multiclass segmentation of digitized kidney tissue sections stained with periodic acid–Schiff (PAS). Methods: We trained the network using multiclass annotations from 40 whole-slide images of stained kidney transplant biopsies and applied it to four independent data sets. We assessed multiclass segmentation performance by calculating Dice coefficients for ten tissue classes on ten transplant biopsies from the Radboud University Medical Center in Nijmegen, The Netherlands, and on ten transplant biopsies from an external center for validation. We also fully segmented 15 nephrectomy samples and calculated the network’s glomerular detection rates and compared network-based measures with visually scored histologic components (Banff classification) in 82 kidney transplant biopsies. Results: The weighted mean Dice coefficients of all classes were 0.80 and 0.84 in ten kidney transplant biopsies from the Radboud center and the external center, respectively. The best segmented class was “glomeruli” in both data sets (Dice coefficients, 0.95 and 0.94, respectively), followed by “tubuli combined” and “interstitium.” The network detected 92.7% of all glomeruli in nephrectomy samples, with 10.4% false positives. In whole transplant biopsies, the mean intraclass correlation coefficient for glomerular counting performed by pathologists versus the network was 0.94. We found significant correlations between visually scored histologic components and network-based measures. Conclusions: This study presents the first convolutional neural network for multiclass segmentation of PAS-stained nephrectomy samples and transplant biopsies. Our network may have utility for quantitative studies involving kidney histopathology across centers and provide opportunities for deep learning applications in routine diagnostics.

1 July 2017 in Laboratory Investigation
Verified
Journal Article
Open access
PDF file
RERenée Mf Ebisch
Renée Mf Ebisch
JHJudith van der Horst
Judith van der Horst
MHMeyke Hermsen
Meyke Hermsen
LRL Lucia Rijstenberg
L Lucia Rijstenberg
JVJudith Em Vedder
Judith Em Vedder
ASAlbert G Siebers
Albert G Siebers
The aim of this study was to evaluate the clinical utility of p16/Ki-67 dual staining, for the identification of CIN in high-risk HPV-positive women from a non-responder screening cohort. P16/Ki-67 dual staining, Pap cytology, and HPV16/18 genotyping were performed on physician-taken liquid-based samples from 495 women who tested high-risk HPV positive on self-sampled material (PROHTECT-3B study). Different triage strategies involving p16/Ki-67 dual staining were evaluated for sensitivity, specificity, and predictive value for ≥CIN2 and ≥CIN3, and compared to Pap cytology with a threshold of atypical cells of undetermined significance. Centrally revised histology or an adjusted endpoint with combined high-risk HPV negative and cytology negative follow-up at 6 months was used as gold standard. Pap cytology (threshold atypical cells of undetermined significance) triage of high-risk HPV-positive samples showed a sensitivity of 93% (95% confidence interval: 85–98) with a specificity of 49% (95% confidence interval: 41–56) for ≥CIN3. Three triage strategies with p16/Ki-67 showed a significantly increased specificity with similar sensitivity. P16/Ki-67 triage of all high-risk HPV-positive samples had a sensitivity of 92% (95% confidence interval: 84–97) and a specificity of 61% (95% confidence interval: 54–69) for ≥CIN3. Applying p16/Ki-67 triage to only high-risk HPV-positive women with low-grade Pap cytology showed a similar sensitivity of 92% (95% confidence interval: 84–97), with a specificity for ≥CIN3 of 64% (95% confidence interval: 56–71). For high-risk HPV-positive women with low-grade and normal Pap cytology, triage with p16/Ki-67 showed a sensitivity of 96% (95% confidence interval: 89–99), and a specificity of 58% (95% confidence interval: 50–65). HPV16/18 genotyping combined with Pap cytology showed a sensitivity and specificity for ≥CIN3 similar to Pap cytology with an atypical cells of undetermined significance threshold. Because the quality of Pap cytology worldwide varies, and differences in sensitivity and specificity are limited between the three selected strategies, p16/Ki-67 triage of all high-risk HPV-positive samples would be the most reliable strategy in triage of high-risk HPV-positive women with an increased specificity and similar sensitivity compared with Pap cytology triage.

16 July 2022 in The American Journal of Pathology
Verified
Journal Article
Open access
MHMeyke Hermsen
Meyke Hermsen
FCFrancesco Ciompi
Francesco Ciompi
AAAdeyemi Adefidipe
Adeyemi Adefidipe
ADAleksandar Denic
Aleksandar Denic
ADAmélie Dendooven
Amélie Dendooven
Jeroen van der LaakJeroen van der Laak
Jeroen van der Laak
In kidney transplant biopsies, both inflammation and chronic changes are important features that predict long-term graft survival. Quantitative scoring of these features is important for transplant diagnostics and kidney research. However, visual scoring is poorly reproducible and labor intensive. The goal of this study was to investigate the potential of convolutional neural networks (CNNs) to quantify inflammation and chronic features in kidney transplant biopsies. A structure segmentation CNN and a lymphocyte detection CNN were applied on 125 whole-slide image pairs of periodic acid-Schiff- and CD3-stained slides. The CNN results were used to quantify healthy and sclerotic glomeruli, interstitial fibrosis, tubular atrophy, and inflammation within both nonatrophic and atrophic tubuli, and in areas of interstitial fibrosis. The computed tissue features showed high correlation with Banff lesion scores of five pathologists (A.A., A.Dend., J.H.B., J.K., and T.N.). Analyses on a small subset showed a moderate correlation toward higher CD3+ cell density within scarred regions and higher CD3+ cell count inside atrophic tubuli correlated with long-term change of estimated glomerular filtration rate. The presented CNNs are valid tools to yield objective quantitative information on glomeruli number, fibrotic tissue, and inflammation within scarred and non-scarred kidney parenchyma in a reproducible manner. CNNs have the potential to improve kidney transplant diagnostics and will benefit the community as a novel method to generate surrogate end points for large-scale clinical studies.