EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multi-Parametric Ultrasound Radiomic Kinetics with Machine Learning Ensemble for Early Prediction of Pathologic Response to Neoadjuvant Chemotherapy in Breast Cancer

Multi-parametrische Ultraschall-Radiomik-Kinetik mit Machine-Learning-Ensemble zur frühen Vorhersage des pathologischen Ansprechens auf neoadjuvante Chemotherapie bei Brustkrebs

Ramona Putin, Livia Stanga, Ciprian Ilie Roșca iD, Horia Silviu Branea iD, Adrian Cosmin Ilie, Alina Tanase, Coralia Cotoraci

Diagnostics · 08.08.2026 · 16 (16) · 2504

DOI: 10.3390/diagnostics16162504 ↗

Auf einen Blick

Bei 135 Frauen mit Brustkrebs sagte ein Ensemble aus klinischen, QUS- und SWE-Verlaufsmerkmalen das pathologische Ansprechen auf NAC früh vorher. Das Modell erreichte eine AUC von 0,93.

Fragestellung

Kann mpUS-Radiomik-Kinetik mit Machine-Learning-Ensemble das pathologische Ansprechen auf NAC bei Brustkrebs früh vorhersagen?

Methode

Prospektive Beobachtungskohorte mit 135 Frauen; QUS- und SWE-Messungen zu Baseline, Woche 1 und 3; nested 10-fold cross-validation, Bootstrap, SHAP und Decision-Curve-Analyse.

Ergebnisse

60 Patientinnen (44,4 %) waren Responder, 75 (55,6 %) Non-Responder. Das Ensemble erreichte eine AUC von 0,93 (95 %-KI 0,88–0,97) vs. 0,71 im klinischen Modell.

Einschränkungen

Alle Leistungsangaben beruhen nur auf interner Kreuzvalidierung; externe Validierung steht aus.

Schlussfolgerung der Autor:innen

Die Autoren schlussfolgern, dass mpUS-Radiomik-Kinetik in einem Machine-Learning-Ensemble einen interpretierbaren Biomarker für das frühe NAC-Ansprechen liefern könnte, vorbehaltlich externer Validierung.

Kernaussagen

  • AUC des Ensembles: 0,93 (95 %-KI 0,88–0,97)
  • Responder zeigten stärkere Veränderungen in MBF, Entropie und SWE-Stiffness
  • Alle Performancewerte stammen nur aus interner Kreuzvalidierung

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Fachartikel in der Zeitschrift Diagnostics, online erschienen am 08.08.2026. Verfasst von Ramona Putin und 6 weiteren Autor:innen. Beteiligte Einrichtung: Doctoral School, Faculty of Medicine, Vasile Goldis Western University of Arad, 310414 Arad, Romania. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Victor Babes University of Medicine and Pharmacy Timisoara.

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background/Objectives: Early identification of breast cancer patients unlikely to benefit from neoadjuvant chemotherapy (NAC) remains a pressing clinical problem because ineffective therapy delays definitive surgery and exposes patients to unnecessary toxicity. Quantitative ultrasound (QUS) and shear wave elastography (SWE) probe complementary tissue properties—scatterer microstructure and mechanical stiffness—that may change before macroscopic tumor shrinkage. This study aimed to evaluate whether multi-parametric ultrasound (mpUS) radiomic kinetics, analyzed with a machine learning ensemble and interpreted with SHAP, could predict pathologic response to NAC.

Methods

A prospective observational cohort enrolled 135 women with biopsy-proven stage II–III breast cancer treated with NAC within the multidisciplinary breast pathway shared between Vasile Goldis Western University of Arad and Victor Babes University of Medicine and Pharmacy Timisoara (Pius Brinzeu County Emergency Hospital). All patients underwent standardized QUS and SWE acquisitions at baseline, week 1, and week 3. Response was defined pathologically at surgery as residual cancer burden (RCB) class 0/I versus II/III. Group comparisons used Welch’s t-test, Mann–Whitney U, chi-square, and Fisher’s exact tests; correlations used Spearman’s rho. A stacked machine learning ensemble (four base learners—XGBoost, random forest, support vector machine, and L2-penalized logistic regression—combined by a separate second-stage logistic meta-learner) was trained with nested 10-fold cross-validation, bootstrap stability assessment, SHAP-based interpretability, and decision curve analysis.

Results

Sixty patients (44.4%) were responders and 75 (55.6%) were non-responders. Responders showed greater week 3 increases in mid-band fit (3.4 ± 0.9 vs. 1.2 ± 0.8 dB, p < 0.001), entropy (0.7 ± 0.2 vs. 0.2 ± 0.2, p < 0.001), and more pronounced SWE mean stiffness reduction (−44.1 ± 9.7 vs. −12.1 ± 8.6 kPa, p < 0.001). The stacked ensemble integrating clinical, QUS, and SWE kinetic features reached an AUC of 0.93 (95% CI 0.88–0.97) versus 0.71 for the clinical-only model (all reported performance figures represent internal cross-validation only). SHAP analysis identified Δ MBF and Δ entropy at week 3 as the dominant features, with high bootstrap stability.

Conclusions

Multi-parametric ultrasound radiomic kinetics integrated through a machine learning ensemble may provide an interpretable early-response biomarker for NAC in breast cancer, pending external validation.

Schlagwörter

Stichwörter der Autor:innen

breast neoplasmselasticity imaging techniquesmachine learningneoadjuvant therapyultrasonography

Autor:innen und Institutionen

  • Doctoral School, Faculty of Medicine, Vasile Goldis Western University of Arad, 310414 Arad, Romania.
  • Discipline of Microbiology, Faculty of Medicine, "Victor Babeș" University of Medicine and Pharmacy, 300041 Timisoara, Romania.
  • Department of Internal Medicine I-Medical Semiotics II, "Victor Babeș" University of Medicine and Pharmacy, 300041 Timisoara, Romania.
  • Department of Functional Sciences, Discipline of Public Health, Center for Translational Research and Systems Medicine, "Victor Babeș" University of Medicine and Pharmacy, 300041 Timisoara, Romania.
  • Department of Professional Legislation in Dental Medicine, Faculty of Dental Medicine, "Victor Babeș" University of Medicine and Pharmacy, Eftimie Murgu Square No. 2, 300041 Timisoara, Romania.
  • Research Centre in Dental Medicine Using Conventional and Alternative Technologies, Faculty of Dental Medicine, "Victor Babeș" University of Medicine and Pharmacy, Eftimie Murgu Square 2, 300041 Timisoara, Romania.

ORCID-Profile: Ciprian Ilie Roșca, Horia Silviu Branea

Förderung

  • Victor Babes University of Medicine and Pharmacy Timisoara

Zitieren

APA

Putin, R., Stanga, L., Roșca, C. I., Branea, H. S., Ilie, A. C., Tanase, A., & Cotoraci, C. (2026). Multi-Parametric Ultrasound Radiomic Kinetics with Machine Learning Ensemble for Early Prediction of Pathologic Response to Neoadjuvant Chemotherapy in Breast Cancer. Diagnostics, 16(16), 2504. https://doi.org/10.3390/diagnostics16162504

Vancouver

Putin R, Stanga L, Roșca CI, Branea HS, Ilie AC, Tanase A, et al. Multi-Parametric Ultrasound Radiomic Kinetics with Machine Learning Ensemble for Early Prediction of Pathologic Response to Neoadjuvant Chemotherapy in Breast Cancer. Diagnostics. 2026;16(16):2504. doi:10.3390/diagnostics16162504

BibTeX
@article{Putin2026,
  author = {Putin, Ramona and Stanga, Livia and Roșca, Ciprian Ilie and Branea, Horia Silviu and Ilie, Adrian Cosmin and Tanase, Alina and Cotoraci, Coralia},
  title = {Multi-Parametric Ultrasound Radiomic Kinetics with Machine Learning Ensemble for Early Prediction of Pathologic Response to Neoadjuvant Chemotherapy in Breast Cancer},
  journal = {Diagnostics},
  year = {2026},
  volume = {16},
  number = {16},
  pages = {2504},
  doi = {10.3390/diagnostics16162504}
}
Kostenlos starten →