Diagnosis of Breast Cancer on Mammography using Attention-Based Convolutional Neural Network
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Breast cancer is the most prevalent cancer in women, and its mortality numbers is rising globally. The reason for the increasing death rate is the finding of breast cancer in malignant stages (last stages). One of the finest methods for detecting breast cancer is mammography. As there are no symptoms at the early stages doctors feel difficult to identify breast cancer. Doctors feel very hard to identify the early stage of breast cancer in mammography films through their naked eye. To address these issues, a methodology for automatically detecting breast cancer in its preliminary stages has been developed. A system built using artificial intelligence has been designed to detect cancer in its early stages utilizing mammography films. Due to the monochromatic character of mammography images, salt and pepper noise as well as Gaussian noise was visible. The presence of the noise gives a grainy appearance to the mammography images. The median filter is used for pre-processing, and its accuracy is evaluated. The performance of the median filter is high, hence the mammography images are filtered using the median filter. To develop an automated method of breast cancer detection the pre-processed image is segmented and classified. Fuzzy C- Means Segmentation (FCM) is used to segment mammography images. Using an Attention Based Convolutional Neural Network (ABCNN), the classification of images is performed. The weights of the affected area are increased whereas the weights of other areas are decreased in an attention-based convolutional neural network. Hence it is easier to classify the affected area in earlier stages of cancer. This ABCNN enhances the classification performance without adding too many network parameters. So the computational cost is low when compared to other methods. The accuracy, specificity, and sensitivity of the classifier are evaluated in order to determine its performance. The combination of Fuzzy C-Means Segmentation and Attention Based Convolutional Neural Network produces high classification accuracy compared to other segmentation methods.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it