Chemotherapy resistance in colorectal cancer (CRC) represents a critical clinical challenge leading to treatment failure and poor patient prognosis. Artemisinin is a natural product isolated from Artemisia annua, and its clinically relevant derivatives include dihydroartemisinin (DHA) and artesunate. Beyond their established antimalarial efficacy, both artemisinin and its derivatives—collectively referred to as artemisinin-derived compounds (ADs)—have been increasingly recognized for their unique potential to reverse multidrug resistance in cancer. Unlike previous reviews focusing on isolated mechanisms, this review systematically constructs a multidimensional, synergistic attack network centered on ADs to elucidate their integrated actions against chemotherapy-resistant CRC. Mechanistically, ADs suppress cancer stem cell (CSC)-associated resistance phenotypes while concurrently reshaping the tumor immune microenvironment, highlighting a functional coupling between stemness inhibition and immune remodeling. In parallel, this review presents apoptosis reactivation and ferroptosis induction as complementary, dual-track cell death strategies that collectively circumvent apoptosis resistance. Moreover, ADs exert “one-strike–multiple-effects” through coordinated regulation of pro-survival signaling networks and immune-related pathways, including the induction of immunogenic cell death (ICD) and the modulation of immunosuppressive macrophage subsets. Beyond mechanistic insights, this review integrates emerging translational considerations, including clinical pharmacokinetics, safety and tolerability, formulation and delivery strategies, and rational combination therapy paradigms in CRC. Collectively, these findings position ADs as multi-dimensional modulators rather than a single-agent cytotoxic, providing a coherent mechanistic and translational rationale for their further development in chemotherapy-resistant CRC.

Rapid and nondestructive assessment of sodium metabisulfite (Na₂S₂O₅) bleaching treatments and sulfur dioxide (SO₂) residues is important for the quality and safety control of Lanzhou lily. This study integrated hyperspectral imaging (HSI), machine learning, and SHapley Additive exPlanations (SHAP) to identify bleaching levels and estimate treatment-associated SO₂ residues. Model performance was assessed using treatment-replicate-grouped validation with samples from a single procurement lot. For bleaching-level classification, the support vector machine (SVM) combined with normalization (NOR), denoted as SVM-NOR, achieved the highest accuracy of 0.902. For SO₂ residue estimation, the random forest (RF) model using raw spectra, denoted as RF-Raw, provided the best performance, with a cross-validated coefficient of determination (R²) of 0.978 and a root mean square error (RMSE) of 38.5 mg/kg. The reported regression performance reflects treatment levels represented in calibration. Complementary leave-one-concentration-out (LOCO) validation yielded a higher pooled RMSE of 155.4 mg/kg for held-out levels. SHAP analysis revealed different wavelength-contribution patterns for classification and regression, providing an interpretable basis for the model predictions. These findings suggest the potential of HSI for rapid prescreening, while further validation is needed for predictions between calibrated treatment levels, particularly near the regulatory reference.