A noninvasive scoring model for the differential diagnosis of ACTH-dependent Cushing’s syndrome: a retrospective analysis of 311 patients based on easy-to-use parameters

Lyu, Xiaohong; Zhang, Dingyue; Pan, Hui; Zhu, Huijuan; Chen, Shi; Lü, Lin

doi:10.1007/s12020-022-03081-0

Cited by 7 publications

(4 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Впервые исследования возможности методов машинного обучения в области дифференциальной диагностики АКТГ-зависимых форм эндогенного гиперкортицизма (БИК и АКТГ-ЭС) проведены китайскими учеными Lyu X. и соавт. [ 13 ][ 14 ]. Данные 311 пациентов, включая 47 больных с АКТГ-ЭС, были использованы для моделирования.…”

Section: Discussionunclassified

“…Данные 311 пациентов, включая 47 больных с АКТГ-ЭС, были использованы для моделирования. На основании многомерного логистического регрессионного анализа из одиннадцати переменных женский пол (отношение шансов (ОШ) — 3,030), гипокалиемия (ОШ — 0,209), уровень АКТГ (ОШ — 0,988), визуализация аденомы гипофиза на МРТ (ОШ — 8,671)и положительный результат БДП (ОШ — 2,768) идентифицированы как предикторы, ассоциированные с БИК, у пациентов с АКТГ-зависимым эндогенным гиперкортицизмом [ 13 ]. Из 8 созданных моделей GBM давала наибольшие значения площади под ROC-кривой (AUC 0,980±0,02), однако модель RF оказалась наилучшей для прогнозирования БИК среди всех моделей машинного обучения.…”

Section: Discussionunclassified

See 1 more Smart Citation

Machine learning methods in differential diagnosis of ACTH-dependent hypercortisolism

Golounina,

Belaya,

Voronov

et al. 2024

Probl Endokrinol (Mosk)

View full text Add to dashboard Cite

AIM: To develop a noninvasive method of differential diagnosis of ACTH-dependent hypercortisolism, as well as to evaluate the effectiveness of an optimal algorithm for predicting the probability of ectopic ACTH syndrome (EAS) obtained using machine learning methods based on the analysis of clinical data.MATERIALS AND METHODS: As part of a single-center, one-stage, cohort study, a retrospective prediction of the probability of EAS among patients with ACTH-dependent hypercortisolism was carried out. Patients were randomly stratified into 2 samples: training (80%) and test (20%). Eleven machine learning algorithms were used to develop predictive models: Linear Discriminant Analysis, Logistic Regression, elastic network (GLMNET), Support Vector machine (SVM Radial), k-nearest neighbors (kNN), Naive Bayes, binary decision tree (CART), C5.0 decision tree algorithms, Bagged CART, Random Forest, Gradient Boosting (Stochastic Gradient Boosting, GBM).RESULTS: The study included 223 patients (163 women, 60 men) with ACTH-dependent hypercortisolism, of which 175 patients with Cushing’s disease (CD), 48 — with EAS. As a result of preliminary data processing and selection of the most informative signs, the final variables for the classification and prediction of EAS were selected: ACTH level at 08:00 hours, potassium level (the minimum value of potassium in the active stage of the disease), 24-h urinary free cortisol, late-night serum cortisol, late-night salivary cortisol, the largest size of pituitary adenoma according to MRI of the brain. The best predictive ability in a training sample of all trained machine learning models for all three final metrics (ROC-AUC (0.867), sensitivity (90%), specificity (56.4%)) demonstrated a model of gradient boosting (Generalized Boosted Modeling, GBM). In the test sample, the AUC, sensitivity and specificity of the model in predicting EAS were 0.920; 77.8% and 97.1%, respectively.CONCLUSION: The prognostic model based on machine learning methods makes it possible to differentiate patients with EAS and CD based on basic clinical results and can be used as a primary screening of patients with ACTH-dependent hypercortisolism.

show abstract

Section: Discussionunclassified

Machine learning methods in differential diagnosis of ACTH-dependent hypercortisolism

Golounina,

Belaya,

Voronov

et al. 2024

Probl Endokrinol (Mosk)

View full text Add to dashboard Cite

show abstract

“…The non-invasive model yielded a higher diagnostic accuracy than HDDST (AUC: 0.915 vs. 0.756) and scores of ≥ −10 and ≥ 13 provided 100 % sensitivity and specificity, respectively, for diagnosis of CD. The authors concluded that BIPSS may only be performed in patients with scores between -10 and 12, as scores below and above these were associated with a high probability of EAS and CD, respectively [16].…”

Section: Discussionmentioning

confidence: 99%

Utility of Simple and Non-Invasive Strategies Alternative to Inferior Petrosal Sinus Sampling and Peripheral CRH Stimulation in Differential Diagnosis of ACTH-Dependent Cushing Syndrome

Attri,

Goyal,

Kalaivani

et al. 2023

Horm Metab Res

View full text Add to dashboard Cite

We aimed to evaluate the utility of simple, cost-effective, and non-invasive strategies alternative to BIPSS and peripheral CRH stimulation in differential diagnosis of ACTH-dependent CS. We performed ROC analysis to evaluate the performance of various tests for differential diagnosis of ACTH-dependent CS in our cohort (CD, n=76 and EAS, n=23) and derived their optimal cut-offs. Subsequently, combining various demographic (gender), clinical (hypokalemia), biochemical (plasma ACTH, HDDST, peripheral CRH stimulation) and imaging (MRI pituitary) parameters, we derived non-invasive models with 100% PPV for CD. Patients with pituitary macroadenoma (n=14) were excluded from the analysis involving non-invasive models. Relative percent ACTH (AUC: 0.933) and cortisol (AUC: 0.975) increase on peripheral CRH stimulation demonstrated excellent accuracy in discriminating CD from EAS. Best cut-offs for CD were plasma ACTH <97.3 pg/mL, HDDST ≥57% cortisol suppression, CRH stimulation ≥77% ACTH increase and ≥11% cortisol increase. We derived six models that provided 100% PPV for CD and precluded the need for BIPPS in 35/85 (41.2%) patients with ACTH-dependent CS and no macroadenoma (in whom BIPSS would have otherwise been recommended). The first three models included basic parameters and avoided both peripheral CRH stimulation and BIPSS in 19 (22.4%) patients, while the next three models included peripheral CRH stimulation and avoided BIPSS in another 16 (18.8%) patients. Using simple and non-invasive alternative strategies, BIPSS can be avoided in 41% and peripheral CRH stimulation in 22% of patients with ACTH-dependent CS and no macroadenoma; such patients can be directly referred for a pituitary surgery.

show abstract

“…However, the profound hypokalemia and rapid progression are clues that often evoke the diagnosis of EAS [13]. In a study using machine learning, the most important features to distinguish EAS from CD were serum potassium, ACTH, and MRI, while other clinical variables, including age, were less important [42].…”

Section: Introductionmentioning

confidence: 99%

Cushing's Syndrome in the Elderly

Zdrojowy-Wełna,

Valassi

2024

Exp Clin Endocrinol Diabetes

View full text Add to dashboard Cite

Management of Cushing’s syndrome (CS) can be particularly challenging in older patients, due to the lack of several clinical features associated with cortisol excess along with a greater burden of comorbidities in them, as compared with the younger. Moreover, interpretation of diagnostic tests could be influenced by age-related physiological changes of cortisol secretion. While mortality is higher and quality of life is more impaired in the elderly with CS as compared with the younger, there is currently no agreement on the most effective therapeutic options in aged individuals, and safety data concerning medical treatment are scanty. In this review, we summarize the current knowledge about age-related differences in etiology, clinical presentation, treatment and outcomes of CS, and describe the potential underlying mechanisms.

show abstract

A noninvasive scoring model for the differential diagnosis of ACTH-dependent Cushing’s syndrome: a retrospective analysis of 311 patients based on easy-to-use parameters

Cited by 7 publications

References 31 publications

Machine learning methods in differential diagnosis of ACTH-dependent hypercortisolism

Machine learning methods in differential diagnosis of ACTH-dependent hypercortisolism

Utility of Simple and Non-Invasive Strategies Alternative to Inferior Petrosal Sinus Sampling and Peripheral CRH Stimulation in Differential Diagnosis of ACTH-Dependent Cushing Syndrome

Cushing's Syndrome in the Elderly

Contact Info

Product

Resources

About