Explainable PET-Based Habitat and Peritumoral Machine Learning Model for Predicting Progression-free Survival in Clinical Stage IA Pure-Solid Non-small Cell Lung Cancer: A Two-center Study

Academic Radiology, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 1, 2025

Language: Английский

---

Quantifying intratumoral heterogeneity within sub-regions to predict high-grade patterns in clinical stage I solid lung adenocarcinoma

BMC Cancer, Journal Year: 2025, Volume and Issue: 25(1)

Published: Jan. 9, 2025

This study aims to quantify intratumoral heterogeneity (ITH) using preoperative CT image and evaluate its ability predict pathological high-grade patterns, specifically micropapillary and/or solid components (MP/S), in patients diagnosed with clinical stage I lung adenocarcinoma (LADC). In this retrospective study, we enrolled 457 who were postoperatively LADC from two medical centers, assigning them either a training set (n = 304) or test 153). Sub-regions within the tumor identified K-means method. Both ecological diversity features (hereafter referred as ITH) conventional radiomics C-radiomics) extracted generate ITH scores C-radiomics scores. Next, univariate multivariate logistic regression analyses employed identify clinical-radiological (Clin-Rad) associated MP/S (+) group for constructing Clin-Rad classification. Subsequently, hybrid model which presented nomogram was developed, integrating classification score. The performance of models assessed receiver operating characteristic (ROC) curves, area under curve (AUC), accuracy, sensitivity, specificity determined. score outperformed both classification, evidenced by higher AUC values (0.820 versus 0.810 0.700, p 0.049 0.031, respectively) (0.805 0.771 0.732, 0.041 0.025, respectively). Finally, consistently demonstrated robust predictive capabilities identifying presence components, achieving 0.830 0.849 (all < 0.05). derived sub-region has been shown be reliable predictor LADC.

Language: Английский

---

Habitat Radiomics Based on Dynamic Contrast‐Enhanced Magnetic Resonance Imaging for Assessing Axillary Lymph Node Burden in Clinical T1–T2 Stage Breast Cancer: A Multicenter and Interpretable Study

Journal of Magnetic Resonance Imaging, Journal Year: 2025, Volume and Issue: unknown

Published: April 21, 2025

ABSTRACT Background Axillary lymph node burden(ALNB) is a critical factor in determining treatment strategies for clinical T 1 –T 2 (cT ) stage breast cancer. However, as ALNB assessment relies on invasive procedures, exploring non‐invasive methods essential. Purpose To develop and validate habitat radiomics model assessing cT cancer, incorporating radiogenomic data to improve interpretability. Study Type Retrospective. Population 468 patients with cancer from two institutions The Cancer Imaging Archive (TCIA) Genome Atlas (TCGA)‐Breast Invasive Carcinoma (BRCA) were included. cohort was divided into training ( n = 173), internal validation 58), external 130), TCGA‐BRCA sets 107). Patients categorized high nodal burden (HNB; > 3 positive nodes) non‐HNB (≤ groups. Field Strength/Sequence 1.5‐T MRI 3.0‐T MRI, three‐dimensional dynamic contrast‐enhanced T1‐weighted gradient‐echo sequences. Assessment Two logistic regression models developed using habitat‐based features. Model performance evaluated the AUC. SHapley Additive exPlanations (SHAP) analysis employed identify key Radiogenomic analysis, including gene set enrichment drug sensitivity assessments, conducted transcriptomic set. Statistical Tests Pearson correlation, Mann–Whitney U , genetic algorithm, regression, AUC delong test, SHAP analysis. A p ‐value < 0.05 considered statistically significant. Results Habitat outperformed Clinical (AUCs: 0.840–0.932 vs. 0.558–0.673). used rank feature importance, subregion showing highest average value. indicated upregulation of KEGG ribosome pathway HNB group identified differential profiles among risk Data Conclusion has potential assess assist radiologists axillary diagnosis, which may help reduce need unnecessary ALN dissection. Evidence Level: 3. Technical Efficacy: Stage 2.

Language: Английский

---

Encoding the Intra- and Peri-Tumoral Heterogeneity of Hepatocellular Carcinoma for Micro-Vascular Invasion Prediction and Prognostic Risk Stratification: A Multicenter Study

Published: Jan. 1, 2025

Language: Английский

---

Computed Tomography‐Based Habitat Analysis for Prognostic Stratification in Colorectal Liver Metastases

Cancer Innovation, Journal Year: 2025, Volume and Issue: 4(2)

Published: March 12, 2025

ABSTRACT Background Colorectal liver metastasis (CRLM) has a poor prognosis, and traditional prognostic models have certain limitations in clinical application. This study aims to evaluate the value of CT‐based habitat analysis CRLM patients compare it with existing provide more evidence for individualized treatment patients. Methods retrospective included 197 whose preoperative contrast‐enhanced CT images corresponding DICOM Segmentation Objects (DSOs) were obtained from The Cancer Imaging Archive (TCIA). Tumor regions segmented, features representing distinct subregions extracted. An unsupervised K‐means clustering algorithm classified tumors into two clusters based on their characteristics. Kaplan–Meier was used overall survival (OS), disease‐free (DFS), liver‐specific DFS. model's predictive performance compared Clinical Risk Score (CRS) Burden (TBS) using concordance index (C‐index), Integrated Brier (IBS), time‐dependent area under curve (AUC). Results model identified patient significant differences OS, DFS, DFS ( p < 0.01). Compared CRS TBS, demonstrated superior accuracy, particularly higher AUC values improved calibration (lower IBS). Conclusions captures spatial tumor heterogeneity provides enhanced stratification CRLM. method outperforms conventional offers potential personalized planning.

Language: Английский

---

Habitat Radiomics for Predicting Progression and Survival in Early-Stage Non-Small Cell Lung Cancer: A Multicenter Study

Published: Jan. 1, 2025

Language: Английский

---

Comparing Habitat, Radiomics and Fusion Models for Predicting Micropapillary/Solid Components in Clinical Stage I Lung Adenocarcinoma: A Multicenter, Retrospective Study

Published: Jan. 1, 2025

Language: Английский

---

Can Habitat-Based MRI Radiomics Distinguish Between T2 and T3 Stages in Rectal Cancer?

Academic Radiology, Journal Year: 2025, Volume and Issue: unknown

Published: April 1, 2025

Language: Английский

---

Interpretable Prognostic Modeling for Postoperative Pancreatic Cancer Using Multi-machine Learning and Habitat Radiomics: A Multi-center Study

Academic Radiology, Journal Year: 2025, Volume and Issue: unknown

Published: May 1, 2025

Language: Английский

---

Habitat-based radiomics from contrast-enhanced CT and clinical data to predict lymph node metastasis in clinical N0 peripheral lung adenocarcinoma ≤ 3 cm

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: May 16, 2025

This study aims to develop an integrated model combining habitat-based radiomics and clinical data predict lymph node metastasis in patients with N0 peripheral lung adenocarcinomas measuring ≤ 3 cm diameter. We retrospectively analyzed 1132 adenocarcinoma from two centers who underwent surgical resection dissection had preoperative computed tomography (CT) scans showing nodules cm. Multivariable logistic regression was employed identify independent risk factors for the model. Radiomics habitat models were constructed by extracting analyzing radiomic features regions contrast-enhanced CT images. Subsequently, a combined developed integrating characteristics. Model performance evaluated using area under receiver operating characteristic curve (AUC). The exhibited promising predictive metastasis, outperforming other standalone AUCs of 0.962, 0.865, 0.853 training, validation, external test cohorts, respectively. demonstrated superior discriminative ability, achieving highest 0.983, 0.950, 0.877 integration offers non-invasive approach assess potentially supporting clinicians optimizing patient management decisions.

Language: Английский

---