MZMichał Ziołkowski

Michał Ziołkowski

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Michał Ziołkowski has a total of 33 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Clinically Actionable Explainable AI in Pulmonary Arterial Hypertension: Endpoints, Calibration, and External Validation. Reply to Pagnoni et al. Toward Clinically Actionable Explainable AI in Pulmonary Arterial Hypertension: Endpoints, Calibration, and External Validation. Comment on “Ledziński et al. Personalized Medicine in Pulmonary Arterial Hypertension: Utilizing Artificial Intelligence for Death Prevention. J. Clin. Med. 2025, 14, 8325”

    by Łukasz Ledziński, Grzegorz Grześk, Michał Ziołkowski, Marcin Waligóra, Marcin Kurzyna, Tatiana Mularek-Kubzdela, Anna Smukowska-Gorynia, Ilona Skoczylas, Łukasz Chrzanowski, Piotr Błaszczak, et al.

    The present Reply addresses the commentary by Pagnoni et al. on our recent study exploring explainable artificial intelligence (AI) for mortality risk prediction in pulmonary arterial hypertension (PAH). We acknowledge the importance of several key issues raised by the authors, including endpoint selection, calibration, decision thresholds, and external validation, all of which are central to translating AI-based prognostic models into clinical practice. Our original endpoint, defined as death by the next follow-up visit, was driven by the structure of nationwide registry data and reflects real-world clinical workflows, although we recognize the advantages of predefined time horizons and time-to-event approaches for future analyses. We discuss the trade-off between sensitivity and precision, emphasizing our deliberate prioritization of minimizing false-negative classifications in high-risk patients, while acknowledging the need for structured clinical pathways to manage false-positive results. We further address calibration and threshold selection, underscoring the necessity of additional clinical studies to support intervention-oriented recommendations. The role of phenotypic determinants and modifiable risk factors in enhancing personalization is highlighted as a key direction for future model development. We reaffirm the value of SHAP-based explainability for improving model transparency, while recognizing the need for continued refinement and clinical validation. Finally, we emphasize the strengths and challenges inherent to registry-based analyses, the importance of external validation, and the need for methodologically sound comparisons with established risk calculators. Overall, this exchange underscores the critical role of interdisciplinary collaboration in advancing clinically actionable and interpretable AI solutions for PAH.

  • Open Access

    Personalized Medicine in Pulmonary Arterial Hypertension: Utilizing Artificial Intelligence for Death Prevention

    by Łukasz Ledziński, Grzegorz Grześk, Michał Ziołkowski, Marcin Waligóra, Marcin Kurzyna, Tatiana Mularek-Kubzdela, Anna Smukowska-Gorynia, Ilona Skoczylas, Łukasz Chrzanowski, Piotr Błaszczak, et al.

    Background/Objectives: Pulmonary arterial hypertension (PAH) is a complex cardiovascular disease with a high burden of morbidity and mortality. Although several risk prediction models have been proposed, the exact significance of distinct clinical parameters in predicting survival in PAH remains unclear. It is important to emphasize that this study does not aim to validate or contradict existing clinical risk assessment calculators provided by the ESC or other scientific societies. Instead, the goal of this research is to identify and rank clinical parameters according to their importance in predicting mortality in PAH patients using machine learning techniques. Methods: Using the Database of Pulmonary Hypertension in the Polish population (BNP-PL) registry, 1755 adult patients with PAH were selected. Feature engineering was conducted using domain knowledge, guided by European Society of Cardiology (ESC) recommendations. Features were reduced using LASSO regression and sequential feature elimination algorithms. A classification model was built using the XGBoost algorithm, utilizing 17 features. The model was tested on a preselected subset of the BNP-PL data. The Shapley Additive Explanations (SHAP) method was used to explain the model’s predictions and to rank feature importance. Results: The model achieved satisfactory results across evaluated metrics, including an area under the curve of 0.767, accuracy of 0.738, specificity of 0.733, and sensitivity of 0.800. SHAP values effectively ranked the features, corroborating the significance of parameters present in the ESC risk stratification tables. Furthermore, local interpretation of results using SHAP enabled individualized assessment of feature importance, enhancing clinical applicability. Conclusions: The proposed artificial intelligence-based model demonstrates satisfactory predictive capability, highlighting the potential of machine learning techniques to support more personalized approaches to the management of PAH patients. This approach offers complementary insights into traditional risk assessment methods, providing clinicians with a novel tool for individualized risk evaluation and decision-making.


  • Open Access

    Diurnal Oscillations of Fibrinolytic Parameters in Patients with Acute Myocardial Infarction and Their Relation to Platelet Reactivity: Preliminary Insights

    by Joanna Boinska, Marek Koziński, Michał Kasprzak, Michał Ziołkowski, Jacek Kubica, Danuta Rość

    There is limited information about diurnal changes in fibrinolysis parameters after acute myocardial infarction (AMI) and their relationship with on-treatment platelet reactivity. The aim of this study was to assess tissue plasminogen activator (t-PA), plasminogen activator inhibitor type-1 (PAI-1), α2-antiplasmin (α2-AP) activity, and plasmin–antiplasmin (PAP) complexes in 30 AMI patients taking dual antiplatelet therapy (DAPT), i.e., acetylsalicylic acid and clopidogrel. Fibrinolytic parameters were assessed at four time points (6 a.m., 10 a.m., 2 p.m., and 7 p.m.) on the third day after AMI using immunoenzymatic methods. Moreover, platelet reactivity was measured using multiple-electrode aggregometry, to assess potential differences in fibrinolytic parameters in low/high on-aspirin platelet reactivity and low/high on-clopidogrel platelet reactivity subgroups of patients. We detected significant diurnal oscillations in t-PA and PAI-1 levels in the whole study group. However, PAP complexes and α2-AP activity were similar at the analyzed time points. Our study reveals a potential impact of DAPT on the time course of fibrinolytic parameters, especially regarding clopidogrel. We suggest the presence of diurnal variations in t-PA and PAI-1 concentrations in AMI patients, with the highest levels midmorning, regardless of platelet reactivity. Significantly elevated levels of PAI-1 during the evening hours in clopidogrel-resistant patients may increase the risk of thrombosis.