# Application Value of Mathematical Models of Diffusion-Weighted Magnetic Resonance Imaging in Differentiating Breast Cancer Lesions.

> **Reliability concern:** This paper was retracted on 2023-06-21. See notice: 10.1155/2023/9787956.

- **DOI:** [10.1155/2021/1481271](https://doi.org/10.1155/2021/1481271)
- **Rating:** Limited (2.4 out of 5)
- **Verdict:** A clinically relevant DWI model comparison in breast imaging whose contradictory statistical reporting and missing diagnostic accuracy metrics undermine its core conclusions.
- **Framework version:** v5.0
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-06-01
- **Evaluation page:** https://www.nabu.science/paper/10.1155%2F2021%2F1481271

## Strengths

- **Clinically relevant diagnostic question addressed** The study targets a genuine clinical need — non-invasive differentiation of benign from malignant breast lesions — comparing three DWI mathematical models that are of active interest in breast radiology.
  - Dimension: contribution
  - Location: Introduction; Abstract Objective
- **Detailed MRI acquisition protocol reported** The methods section provides specific parameters for MR plain scan, single b-value DWI, multi-b-value DWI (12 b-values specified), and dynamic contrast-enhanced MR, enabling replication of the imaging protocol.
  - Dimension: methodological rigour
  - Location: Methods Section 2.2
- **Systematic comparison across lesion types** The study systematically compares DWI parameters across breast cysts, benign tumours, and malignant tumours against normal gland tissue, and reports ROC-based diagnostic performance for each lesion category.
  - Dimension: contribution
  - Location: Results Sections 3.1–3.8

## Limitations

- **Self-contradictory statistical reporting in results** Sections 3.1 and 3.3 simultaneously describe DWI parameters as 'not different from those of normal glands' and as having 'statistically significant' differences (P < 0.05). This logical contradiction makes it impossible to determine the actual findings for several key parameters.
  - Dimension: methodological rigour
  - Location: Results Sections 3.1, 3.3
- **Missing AUC values and incomplete ROC reporting** No AUC values, confidence intervals, or decision thresholds are reported for any of the ROC analyses across Sections 3.2, 3.4, 3.6, and 3.8. Without AUC values, the central claim that DDC has the 'most significant diagnostic effect' cannot be evaluated or verified.
  - Dimension: methodological rigour
  - Location: Results Sections 3.2, 3.4, 3.6, 3.8
- **No multiple comparison correction applied** Approximately 40+ independent-samples T-tests are conducted across five DWI parameters and multiple lesion type comparisons with no correction for multiple comparisons, leaving the false discovery rate uncontrolled at the P < 0.05 threshold.
  - Dimension: methodological rigour
  - Location: Methods Section 2.3; Results Sections 3.1–3.7

## Reliability checks

Concerns (4 of 4 checks).

- **Internal coherence:** Concern. Multiple instances of logically contradictory statistical reporting appear in the results sections, where parameters described as 'not different' from normal glands are simultaneously reported as having 'statistically significant' differences (P < 0.05). This pattern recurs across Sections 3.1, 3.3, and potentially 3.5, suggesting systematic misreporting rather than isolated error.
- **Research conduct:** Noted. Ethics approval and informed consent are declared but unverifiable: no committee name or approval identifier is provided. No conflicts of interest statement is present. Data availability is claimed only upon request with no repository or persistent identifier.
- **Reference integrity:** Clean. All 26 cited references resolved to real publications via PubMed, Crossref, or OpenAlex.
  - References not verified automatically are resolved manually within 24 hours.

## Post-publication record

- **Status:** Concern
- **Summary:** This paper was retracted on 2023-06-21. See notice: 10.1155/2023/9787956.
- **Last checked:** 2026-10-17
