Lisa Maria Hillen

Lisa Maria Hillen

Auto-generated profile, based on publicly available information

Lisa Maria Hillen

Department of Pathology, GROW-Research Institute for Oncology and Reproduction, Maastricht University Medical Centre, Maastricht, the Netherlands

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
16 December 2023 in Medical Image Analysis
Verified
Journal Article
Open access
DGDaan J. Geijs
Daan J. Geijs
SDStephan Dooper
Stephan Dooper
WAWitali Aswolinskiy
Witali Aswolinskiy
LHLisa M. Hillen
Lisa M. Hillen
AAAvital L. Amir
Avital L. Amir
GLGeert Litjens
Geert Litjens
The frequency of basal cell carcinoma (BCC) cases is putting an increasing strain on dermatopathologists. BCC is the most common type of skin cancer, and its incidence is increasing rapidly worldwide. AI can play a significant role in reducing the time and effort required for BCC diagnostics and thus improve the overall efficiency of the process. To train such an AI system in a fully-supervised fashion however, would require a large amount of pixel-level annotation by already strained dermatopathologists. Therefore, in this study, our primary objective was to develop a weakly-supervised for the identification of basal cell carcinoma (BCC) and the stratification of BCC into low-risk and high-risk categories within histopathology whole-slide images (WSI). We compared Clustering-constrained Attention Multiple instance learning (CLAM) with StreamingCLAM and hypothesized that the latter would be the superior approach. A total of 5147 images were used to train and validate the models, which were subsequently tested on an internal set of 949 images and an external set of 183 images. The labels for training were automatically extracted from free-text pathology reports using a rule-based approach. All data has been made available through the COBRA dataset. The results showed that both the CLAM and StreamingCLAM models achieved high performance for the detection of BCC, with an area under the ROC curve (AUC) of 0.994 and 0.997, respectively, on the internal test set and 0.983 and 0.993 on the external dataset. Furthermore, the models performed well on risk stratification, with AUC values of 0.912 and 0.931, respectively, on the internal set, and 0.851 and 0.883 on the external set. In every single metric the StreamingCLAM model outperformed the CLAM model or is on par. The performance of both models was comparable to that of two pathologists who scored 240 BCC positive slides. Additionally, in the public test set, StreamingCLAM demonstrated a comparable AUC of 0.958, markedly superior to CLAM's 0.803. This difference was statistically significant and emphasized the strength and better adaptability of the StreamingCLAM approach.

28 February 2023 in Journal of the European Academy of Dermatology and Venereology
Verified
Journal Article
Open access
YWYousif Ahmed WIDAATALLA
Yousif Ahmed WIDAATALLA
TWT. Wolswijk
T. Wolswijk
FAF. Adan
F. Adan
LHL. M. Hillen
L. M. Hillen
Henry C WoodruffHenry C Woodruff
Henry C Woodruff
KMKlara Mosterd
Klara Mosterd
Basal cell carcinoma (BCC) is one of the most common types of cancer. The growing incidence worldwide and the need for fast, reliable and less invasive diagnostic techniques make a strong case for the application of different artificial intelligence techniques for detecting and classifying BCC and its subtypes. We report on the current evidence regarding the application of handcrafted and deep radiomics models used for the detection and classification of BCC in dermoscopy, optical coherence tomography and reflectance confocal microscopy. We reviewed all the articles that were published in the last 10 years in PubMed, Web of Science and EMBASE, and we found 15 articles that met the inclusion criteria. We included articles that are original, written in English, focussing on automated BCC detection in our target modalities and published within the last 10 years in the field of dermatology. The outcomes from the selected publications are presented in three categories depending on the imaging modality and to allow for comparison. The majority of articles (n = 12) presented different AI solutions for the detection and/or classification of BCC in dermoscopy images. The rest of the publications presented AI solutions in OCT images (n = 2) and RCM (n = 1). In addition, we provide future directions for the application of these techniques for the detection of BCC. In conclusion, the reviewed publications demonstrate the potential benefit of AI in the detection of BCC in dermoscopy, OCT and RCM.

9 January 2018 in Frontiers in Microbiology
Verified
Journal Article
Open access
LHLisa M. Hillen
Lisa M. Hillen
DRDorit Rennspiess
Dorit Rennspiess
ESErnst Jan Maria Speel
Ernst Jan Maria Speel
AHAnke M. Haugg
Anke M. Haugg
VWVéronique Winnepenninckx
Véronique Winnepenninckx
AHAxel zur Hausen
Axel zur Hausen
Seborrheic keratosis (SK) is the most common benign cutaneous neoplasm. A subset shows increased p16 expression. Since SK shares several features with verruca vulgaris, e.g., increased p16 expression, human papillomaviruses (HPV) have been suggested as possible causal agents. However, a relevant association could not be established between HPV and SK. In the present study we aimed to investigate the presence of Merkel cell polyomavirus (MCPyV) in relation to p16 expression in SK. P16 expression was investigated using immunohistochemistry (IHC). Presence of MCPyV was assessed in 23 formalin-fixed paraffin-embedded tissue samples of SK by molecular techniques (i.e., PCR and FISH) and IHC. 16/23 SK showed strong to moderate p16 expression. 6/23 of SK were MCPyV positive by PCR which was confirmed by FISH. Of interest, two samples with strong FISH signals also showed MCPyV expression as tested by IHC. Samples with weaker signal intensity were negative in IHC. P16 expression was not associated with the presence of MCPyV. Concluding, the detection of MCPyV DNA by PCR and FISH in SK reflects the widespread prevalence of MCPyV in the skin. However, low detection rates exclude MCPyV as a major pathogenic factor in SK, most likely representing a coincidental infection. P16 IHC does not appear as useful adjunctive surrogate marker for the presence of MCPyV in SK.

25 January 2024 in Molecular Oncology
Verified
Journal Article
Open access
FBFranziska Böttger
Franziska Böttger
TRTeodora Radonic
Teodora Radonic
IBIdris Bahce
Idris Bahce
KMKim Monkhorst
Kim Monkhorst
SPSander R. Piersma
Sander R. Piersma
CJConnie R Jimenez
Connie R Jimenez
The majority of patients with resected stage II‐IIIA non‐small cell lung cancer (NSCLC) are treated with platinum‐based adjuvant chemotherapy (ACT) in a one‐size‐fits‐all approach. However, a significant number of patients do not derive clinical benefit, and no predictive patient selection biomarker is currently available. Using mass spectrometry‐based proteomics, we have profiled tumour resection material of 2 independent, multi‐centre cohorts of in total 67 patients with NSCLC who underwent ACT. Unsupervised cluster analysis of both cohorts revealed a poor response/survival sub‐cluster composed of ~ 25% of the patients, that displayed a strong epithelial‐mesenchymal transition signature and stromal phenotype. Beyond this stromal sub‐population, we identified and validated platinum response prediction biomarker candidates involved in pathways relevant to the mechanism of action of platinum drugs, such as DNA damage repair, as well as less anticipated processes such as those related to the regulation of actin cytoskeleton. Integration with pre‐clinical proteomics data supported a role for several of these candidate proteins in platinum response prediction. Validation of one of the candidates (HMGB1) in a third independent patient cohort using immunohistochemistry highlights the potential of translating these proteomics results to clinical practice.

30 January 2023 in Histopathology
Verified
Journal Article
Open access
AKAnke Keijbeck
Anke Keijbeck
ARAnniek Raaijmaakers
Anniek Raaijmaakers
LHLisa Hillen
Lisa Hillen
MGMarielle Gelens
Marielle Gelens
SKSander van Kuijk
Sander van Kuijk
MCMaarten Christiaans
Maarten Christiaans
Aims: In current renal transplant pathology practice, interstitial fibrosis is visually assessed in categories according to the Banff classification. As this has a moderate reproducibility, which is little ameliorated by morphometric analysis, we investigated whether visual renal fibrosis assessment is feasible on a continuous scale, i.e. as a percentage of affected area of the cortex.Methods and results: Protocol renal biopsies taken at transplantation (n = 125), three (n = 73) and 12 months (n = 88) after transplantation were visually scored in categories (Banff) and percentages for interstitial fibrosis (ci). Interobserver variation (ICC and weighted κ) was assessed, and morphometric analysis on Sirius red‐stained sections was performed. Correlations between the different methods and their association with donor age and eGFR 1 and 5 years post‐transplant were analysed using Pearson's or Spearman's rho. Interobserver agreement was equivalent for Banff and %ci (κ = 0.713 versus ICC = 0.792), and for Banff IF/TA and %IF/TA (κ = 0.615 versus ICC = 0.743). Both Banff and %ci were associated with Sirius red morphometry in 3 and 12 months. With all three methods, a significant correlation was found between donor age and fibrosis in the implantation biopsy and between fibrosis in the 12 months’ biopsy and eGFR at 1 and 5 years (eGFR at 1 year: Sirius red ρ = 0.487, %ci ρ = 0.393, Banff ρ = 0.413, all P < 0.01, eGFR at 5 years: Sirius red ρ = 0.392, %ci ρ = 0.333, Banff ρ = 0.435, all P < 0.01).Conclusion: Interstitial fibrosis assessment on a continuous scale can be used next to scoring in categories according to the Banff classification in protocol renal transplant biopsies.

1 September 2025 in Laboratory Investigation
Verified
Journal Article
Open access
FHFrank W.J. Heijboer
Frank W.J. Heijboer
JDJules L. Derks
Jules L. Derks
DMDana A.M. Mustafa
Dana A.M. Mustafa
NRNicole Rijnsburger
Nicole Rijnsburger
BHBregtje C.M. Hermans
Bregtje C.M. Hermans
JTJan H von der Thüsen
Jan H von der Thüsen
Large-cell neuroendocrine carcinoma (LCNEC) can be genomically subtyped into small-cell lung cancer (SCLC) and non-SCLC-like. Neurogenic differentiation 1 (NEUROD1), achaete-scute homolog 1 (ASCL1), POU class 2 homeobox 3 (POU2F3), and yes-associated protein 1 (YAP1) (NEUROD1, ASCL1, POU2F3, and YAP1 [NAPY]) subtypes have been reported for SCLC. We immunohistochemically evaluated NAPY in LCNEC alongside relevant protein expression data. Tissue microarrays from 133 stage I-III resected LCNEC were reviewed and immunostained for NAPY, protein retinoblastoma (pRb), delta-like ligand 3 (DLL3), cMYC, and thyroid transcription factor 1. An H-score of >10 was considered positive (+), and >50, dominant. Unsupervised clustering and spatial immune RNA profiling using GeoMx Digital Spatial Profiling (NanoString Technology) were performed. Clinical data were obtained from the Netherlands Cancer Registry. ASCL1 was dominant in 26% and NEUROD1 in 18% of LCNEC. pRb loss was observed in 75%, and DLL3+, cMYC+, and thyroid transcription factor 1+ in 66%, 26%, and 70%, respectively. Unsupervised clustering identified 5 expression-based subgroups: NEUROD1high-ASCL1high (10%), ASCL1high (22%), POU2F3high (5%), YAP1high (11%), and NAPYlow (51%). Both ASCL1high subgroups correlated with DLL3high and high neuroendocrine (NE) marker expression. YAP1high was enriched for pRb+. POU2F3high and YAP1high subgroups were NE marker low and DLL3low. GeoMX Digital Spatial Profiling identified 4 upregulated genes involved in immune system and/or tumor development in the NEUROD1high-ASCL1high-POU2F3high- group. In this comprehensive evaluation of NAPY markers in LCNEC, we observed 5 expression-based subgroups: NEUROD1high-ASCL1high, ASCL1high, POU2F3high, YAP1high, and NAPYlow. The NE subgroups (NEUROD1high-ASCL1high and ASCL1high) were recognized with DLL3high expression. Compared with the proportion known in SCLC, more NAPYlow and YAP1high and fewer POU2F3high cases were identified. Application of NAPY in LCNEC provides a more modest discrimination of subgroups compared with SCLC. Further research on potential drug targets is warranted, ie, differences in DLL3 and YAP1 expression could guide personalized treatment strategies.