Ferroptosis, an iron-dependent regulated cell death driven by lipid peroxidation, contributes to intervertebral disc degeneration (IVDD), yet most reviews describe its pathways without asking why the disc—the largest avascular tissue, combining hypoxia, nutrient deprivation, mechanical loading, acidosis, and low-grade inflammation—regulates ferroptosis differently from other tissues. Here we address this question along three axes. First, the execution machinery shows three IVDD-specific vulnerabilities: a proposed diffusion-limited cystine “semi-starvation”, mechanically induced GPX4 inhibition via the Piezo1–Ca2+ axis, and epigenetic GCLC suppression. Second, iron homeostasis undergoes a “triangular imbalance”: TFR1 upregulation coexists with restricted iron supply, acidosis-activated NCOA4 ferritinophagy is proposed as the dominant source of labile iron (with the UCHL3–HMGB1–NCOA4 axis as a newly identified upstream regulator), and FPN1 export is blocked by the IL-6–hepcidin axis, predisposing Nrf2–HO-1 to a protective-to-deleterious functional reversal. Third, inflammation and ferroptosis form a vicious cycle in which STAT3 signals through the non-canonical S100A6–p53 axis—opposite to its tumour-protective FTH1 role—jointly imposing a “dual lock” on SLC7A11. Integrating these findings, we construct a network in which Nrf2 is proposed as the transcriptional hub and SLC7A11 as the convergent bottleneck, and derive an SLC7A11-restoration hypothesis predicting that multi-target combinations, not single-pathway interventions, will be required—a prediction that remains to be tested. We close by discussing the controversies and a prioritized research agenda.
Purpose: Lumbar spine MRI interpretation requires 25 structured severity judgments per examination — five degenerative conditions at five disc levels, each on a three-level ordinal scale — a time-consuming task with substantial inter-reader variability. We developed and externally validated a model that automates this complete scheme. Materials and Methods: A full-sequence multi-instance learning (MIL) model — ResNet34 backbone, bidirectional GRU, attention pooling over 24 uniformly sampled sagittal T2-weighted slices — with 25 cumulative-logit ordinal heads was trained on 1,204 examinations from the multi-institutional RSNA 2024 LumbarDISC dataset and internally validated on 213 held-out examinations. External validation used 60 retrospectively collected patients from an independent Chinese university hospital (four scanner vendors, radiology reports as the reference standard, no fine-tuning). Agreement was assessed with exact accuracy, sensitivity, specificity, McNemar tests, and patient-clustered bootstrap confidence intervals. Results: The ordinal model achieved 51.4% balanced accuracy internally (exact accuracy 65.1%), versus 16.3% for a single-slice baseline; severe-grade AUCs averaged 0.908 across conditions. Externally, it outperformed its softmax-weighted counterpart on exact accuracy (0.649 vs. 0.586; +6.3 percentage points, 95% CI +2.7 to +10.0) and sensitivity (0.868 vs. 0.833), with markedly fewer severe-grade false alarms (69 vs. 172 predicted severe cells; McNemar p<0.0001). Conclusion: A full-sequence MIL model with ordinal outputs automates the complete 25-task grading scheme on sagittal MRI and, in the first hospital external validation of such a model, reduced severe-grade over-prediction under domain shift without loss of sensitivity — supporting its role as a screening and triage aid. Summary: An ordinal full-sequence multi-instance learning model graded all 25 lumbar MRI degeneration tasks and markedly reduced severe-grade false alarms in 60-patient hospital validation without loss of sensitivity. Key Points: Full-sequence attention was indispensable: balanced accuracy rose from 16.3% (single-slice baseline) to 51.4% over 24 sagittal slices. In external validation, ordinal outputs reduced severe-grade predictions from 172 to 69 of 900 cells (McNemar p<0.0001) with comparable ranking ability (AUC 0.835 vs. 0.829). The Moderate grade remained the hardest class across all conditions (AUC 0.70–0.78), tracking known inter-reader variability at the mild–moderate boundary.