CT-based Deep Learning Model For Assessing The Severity Of Patients With Connective Tissue Disease-associated Interstitial Lung Disease

Objective:To develop a CT-based deep learning model for assessing the severity of patients with connective tissue disease-associated interstitial lung disease.Methods:The retrospective study included 298 CTD-ILD patients between January 2018 and May 2022.Chest HRCT images of 114 patients were used to establish a deep learning-based RDNet model(1610 fully annotated CT images for training and 402 images for validation).Meanwhile,a dataset of the remaining 184 CTD-ILD patients was used to evaluate the model(staged according to the gender-age-physiology(GAP)system).The test cohort included 113 males and 71 females,with a median age of 62 years.The model was used to automatically classify and quantify three radiographic features(GGOs,reticulation,and honeycombing),along with a volumetric sum of three areas(ILD%).Meanwhile,the quantitative assessment index was calculated using four previously defined CT thresholds as control(different lung attenuation ranges),and compared with the results of RDNet model analysis.The Spearman rank correlation coefficient(r)was calculated to evaluate the correlation between variables.Results:The RDNet model accurately identified GGOs,reticulation,and honeycombing,with corresponding Dice indexes of 0.784,0.782,and 0.747,respectively.A total of 137 patients were at GAP1(73.9%),36 patients at GAP2(19.6%),and 11 patients at GAP3(6.0%).The percentage of reticulation and honeycombing at GAP2 and GAP3 were markedly elevated compared with those at GAP1(P<0.001).The percentage of GGOs was not significantly different among the GAP stages(P=0.62).As the GAP stage increased,all lung function indicators tended to decrease,and the CPI index(Composite Physiologic Index)indicated an upward tendency.The percentage of honeycombs moderately correlated with DLco%(r=-0.58,P<0.001)and CPI(r=0.63,P<0.001).The ILD assessment index calculated by the CT threshold method(-260 to-600HU)had a low correlation with DLco% and CPI(DLco%: r=-0.42,P<0.001;CPI: r=0.45,P<0.001).Conclusion:The RDNet model can quantify GGOs,reticulation,and honeycombing of chest CT images in CTD-ILD patients,among which honeycombing had the most significant effect on lung function indicators.Additionally,this model provided good clinical utility for evaluating the severity of CTD-ILD.