Arkhiv patologii · 2024
[The algorithm for morphological assessment of malignant potential of adrenocortical tumors using mathematical modeling method].
Аннотация на английскомСтатьяНазвание и аннотация: Оригинал на русском
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Urusova L S, Pachuashvili N V, Porubayeva E E, Elfimova A R, Beltsevich D G, Chevais A, Demura T A, Mokrysheva N G
Аннотация
OBJECTIVE: To develop the mathematical model with high sensitivity and specificity to assess the malignant potential of adrenal cortical tumors, which can be used to diagnose adrenocortical carcinoma (ACC) in adults.
MATERIAL AND METHODS: Pathomorphological examination of surgical and consultative material of adrenocortical neoplasms was carried out. All cases were verified according to the WHO Classification of adrenal gland tumors (5th ed., 2022), the tumor's histogenesis was confirmed by immunohistochemical examination. Statistical analysis of the histological and immunohistochemical factors in terms of their value in relation to the diagnosis of ACC was carried out on Python 3.1 in the Google Colab environment. ROC analysis was used to identify critical values of predictors. The cut-off point was selected according to the Youden`s index. Logistic regression analysis using l1-regularisation was performed. To validate the model, the initial sample was divided into training and test groups in the ratio of 9:1, respectively.
RESULTS: The study included 143 patients divided into training (128 patients) and test (15 patients) samples. A prognostic algorithm was developed, which represent a diagnostically significant set of indicators of the currently used Weiss scale. The diagnosis is carried out in 3 stages. This mathematical model showed 100% accuracy (95% CI: 96-100%) on the training and test samples.
CONCLUSION: The developed algorithm could solve the problem of subjectivity and complexity in the interpretation of some of the criteria of current diagnostic algorithms. The new model is unique in that, unlike others, it allows verification of all morphological variants of ACC.
MATERIAL AND METHODS: Pathomorphological examination of surgical and consultative material of adrenocortical neoplasms was carried out. All cases were verified according to the WHO Classification of adrenal gland tumors (5th ed., 2022), the tumor's histogenesis was confirmed by immunohistochemical examination. Statistical analysis of the histological and immunohistochemical factors in terms of their value in relation to the diagnosis of ACC was carried out on Python 3.1 in the Google Colab environment. ROC analysis was used to identify critical values of predictors. The cut-off point was selected according to the Youden`s index. Logistic regression analysis using l1-regularisation was performed. To validate the model, the initial sample was divided into training and test groups in the ratio of 9:1, respectively.
RESULTS: The study included 143 patients divided into training (128 patients) and test (15 patients) samples. A prognostic algorithm was developed, which represent a diagnostically significant set of indicators of the currently used Weiss scale. The diagnosis is carried out in 3 stages. This mathematical model showed 100% accuracy (95% CI: 96-100%) on the training and test samples.
CONCLUSION: The developed algorithm could solve the problem of subjectivity and complexity in the interpretation of some of the criteria of current diagnostic algorithms. The new model is unique in that, unlike others, it allows verification of all morphological variants of ACC.