{{ toolbarErrorMessage }}
{{ section.description }}
Overall information regarding the model:
The model was based on 2,892 cases of DCIS and 127 events of lymph node metastases. The predicted risks in our study ranged from 0.4% to 40.4%, the mean was 4.4% and the median was 2.8%. The c-index was 0.745 and it was 0.748 after correction for optimism by bootstrapping. In this study the sensitivity was the rate of metastasis that was correctly predicted as high-risk and 1-specificity was the rate of no-metastasis that was falsely predicted as high-risk. The calibration plot had a slope of 1.029 and an intercept of 0.090. The decision curve analysis showed that the model has a higher benefit than performing the sentinel lymph node biopsy for all patients. Below the prediction cut-off point of 25% the model has a higher benefit than performing the sentinel lymph node biopsy for none patients.
The validation cohort consisted of the 2,269 patients. In total, 53 (2.2%) had metastasis. The calibration plot showed that the observed proportions were lower than the predicted proportions. The AUC in the validation cohort was 0.741 (95% confidence interval 0.662 to 0.820). The maximum value of the Youden’s index was found at a predicted risk of 3.5%, with a sensitivity of 71.4% and a specificity of 70.6%.
Patients with previous ipsilateral DCIS or invasive breast cancer, patients with biopsy-proven micro-invasive cancer and patients who underwent excisional biopsy were excluded from the cohorts on which the model was developed or validated.
How to use the model:
The model can be used to calculate the individual risk of lymph node metastasis for patients with biopsy-proven DCIS.
Ce algorithme est fourni à des fins d'éducation, de formation et d'information. Il ne doit pas être utilisé pour aider à la prise de décision médicale ou pour fournir des services médicaux ou de diagnostic. Lire l'intégralité de notre disclaimer.
With an Evidencio Community account you can:
A personal Evidencio account is free, with no strings attached!
Join us and help create clarity, transparency, and efficiency in the creation, validation, and use of medical prediction algorithms.
| {{ (typeof row === 'object') ? row.label : row }} |
| {{ column }} | |
|---|---|
| {{ row.label }} | {{ value }} |
{{ error }}
{{ error }}
Veuillez saisir un mot de passe
Un mot de passe doit comporter au moins 8 caractères
Un mot de passe ne peut pas être plus long que 64 caractères.
Choisissez un mot de passe avec au moins une lettre majuscule
Choisissez un mot de passe avec au moins un caractère spécial (@$!%*#?&)
Veuillez accepter les conditions générales et la clause de non-responsabilité
Veuillez fournir votre adresse e-mail et nous vous enverrons un lien pour réinitialiser votre mot de passe.
Adresse email
Veuillez entrer un email valide
Si un compte a été enregistré avec cette adresse e-mail, vous recevrez un lien de récupération dans votre courrier
Veuillez utiliser le lien de réinitialisation du mot de passe contenu dans le mail pour définir votre nouveau mot de passe
Vous n'avez pas encore reçu l'e-mail ? Veuillez vérifier votre dossier spam, ou renvoyer l'email