Automated inversion time selection for late gadolinium-enhanced cardiac magnetic resonance imaging.

OBJECTIVES: To develop and share a deep learning method that can accurately identify optimal inversion time (TI) from multi-vendor, multi-institutional and multi-field strength inversion scout (TI scout) sequences for late gadolinium enhancement cardiac MRI. MATERIALS AND METHODS: Retrospective multicentre study conducted on 1136 1.5-T and 3-T cardiac MRI examinations from four centres and three scanner vendors. Deep learning models, comprising a convolutional neural network (CNN) that provides input to a long short-term memory (LSTM) network, were trained on TI scout pixel data from centres 1 to 3 to identify optimal TI, using ground truth annotations by two readers. Accuracy within 50 ms, mean absolute error (MAE), Lin's concordance coefficient (LCCC) and reduced major axis regression (RMAR) were used to select the best model from validation results, and applied to holdout test data. Robustness of the best-performing model was also tested on imaging data from centre 4. RESULTS: The best model (SE-ResNet18-LSTM) produced accuracy of 96.1%, MAE 22.9 ms and LCCC 0.47 compared to ground truth on the holdout test set and accuracy of 97.3%, MAE 15.2 ms and LCCC 0.64 when tested on unseen external (centre 4) data. Differences in vendor performance were observed, with greatest accuracy for the most commonly represented vendor in the training data. CONCLUSION: A deep learning model was developed that can identify optimal inversion time from TI scout images on multi-vendor data with high accuracy, including on previously unseen external data. We make this model available to the scientific community for further assessment or development. CLINICAL RELEVANCE STATEMENT: A robust automated inversion time selection tool for late gadolinium-enhanced imaging allows for reproducible and efficient cross-vendor inversion time selection. KEY POINTS: • A model comprising convolutional and recurrent neural networks was developed to extract optimal TI from TI scout images. • Model accuracy within 50 ms of ground truth on multi-vendor holdout and external data of 96.1% and 97.3% respectively was achieved. • This model could improve workflow efficiency and standardise optimal TI selection for consistent LGE imaging.

Skin type specific photobiological response to visible light is mediated by constitutional melanin.

BACKGROUND: Visible light (VL) induces varying photobiological responses between skin types, likely influenced by inherent melanization. Individual typology angle (ITA) objectively measures skin types. We hypothesize that epidermal melanin content and distribution determine VL response. OBJECTIVES: This study describes clinical and histologic responses to VL and examines the potential role of melanin in the underlying mechanistic pathways. METHODS: We grouped enrolled participants by ITA (Light = 5, Intermediate = 4, Dark = 7) per colorimetry (CR-400, Konica Minolta). Photoprotected sites were exposed daily to 480 J/cm2 of VL (Fiber-Lite High Intensity Illuminator, Series 180, Dolan Jenner Industries Inc.) for 4 days (total = 1920 J/cm2 ), as tolerated. Treated and control sites were biopsied 96 h after first exposure. We used hematoxylin and eosin and Fontana-Mason to assess histological changes and melanin deposition, respectively. p53 and Ki67 immunohistochemical stains were done to assess DNA damage and proliferation. Matrix metalloproteinase (MMP)-1 expression was detected by immunohistochemical staining and immunofluorescence microscopy. RESULTS: Darker skin did not tolerate the full VL regimen with blistering occurring in most subjects at doses of 220-880 J/cm2 . Intermediate and Dark skin showed tanning. Light skin developed erythema. p53 counts were highest in Intermediate, followed by Light skin, although this was not statistically significant. VL treatment led to MMP-1 expression and nuclear localization in keratinocytes in Dark and Intermediate but not in Light skin, however differences between groups were not statistically significant. CONCLUSIONS: Skin types demonstrate unique biological responses to VL. The role of melanin in photoprotection is well-defined. However, given the pro-apoptotic function of nuclear MMPs, we suggest a potential mechanism by which melanin may mediate VL-induced phototoxicity.

Periocular Dark Circles: Correlates of Severity.

BACKGROUND: Periocular dark circles (PDCs) are a common cosmetic complaint. Grading systems based on objective measures have been used but no standard system is in place. OBJECTIVE: To determine factors associated with subjective and objective PDC severity. METHODS: Enrolled patients (n=100) completed a questionnaire comprised of demographic variables, medical history, and self-perception of PDC. Those perceiving PDC graded dissatisfaction on a 10-point scale. Clinical severity (grades 0~4) and subtype (constitutional, post-inflammatory, vascular, shadow effects, or others) were determined. A Konica Minolta CR-400 chromameter was used to obtain colorimetry measurements (L*a*b* values). The objective average difference in darkness (ΔL*) between the periocular region and the cheek was determined. Comparisons were made using Spearman correlation coefficients (r). RESULTS: Patient dissatisfaction correlated with both clinical severity (r=0.46, p<0.001) and the ΔL* by colorimetry (r=0.35, p=0.004). Factors associated with subjective dissatisfaction were female sex (r=0.38, p=0.002), higher Fitzpatrick skin type (r=0.42, p=0.001), fewer hours of sleep (r=-0.28, p=0.03), and use of concealer (r=0.35, p=0.004). Factors associated with objective measures were higher Fitzpatrick skin type (r=0.36, p=0.0007 and r=0.28, p=0.009, respectively), family history of PDC (r=0.34, p<0.001 and r=0.20, p=0.05), and history of eczema (r=0.45, p<0.001 and r=0.20, p=0.0504). Clinical severity grading correlated with colorimetric severity (r=0.36, p=0.0003). CONCLUSION: Overall, subjective dissatisfaction was associated with clinical severity. However, factors associated with subjective severity did not necessarily overlap with factors associated with objective severity. These findings highlight the importance of patient-reported grading. There may be added value in incorporating a component of subjective grading into the traditionally objective PDC grading scales.