Deep Learning Mobile Algorithms for Detection of Skin Cancer
Bibliographic record
Abstract
Skin cancer malignant melanoma is the deadliest type of cancer and early detection is important to improve patient prognosis.Recently, Deep Learning Neural Networks (DLNNs) have proven to be a powerful tool in classifying medical images for detecting various diseases and it has become viable to deal with skin cancer detection.In this research we propose a serverless mobile app to assist with skin cancer detection.This mobile app is based on the best performance of five Convolution Neural Network (CNN) models designed from scratch as well as four state-of-the-art architectures used for Transfer Learning (Inception v3, ResNet50v2, DenseNet, and Exception v2).Since the skin cancer dataset is imbalanced, we perform data augmentation.We also use the fine-tuning top layers technique for feature extraction on all models to improve the results.The main novelty of the proposed method is the model deployed as part of mobile app where the classification processes are executed locally on the mobile device.This approach will reduce the latency and improve the privacy of the end users compared with the cloud-based model where user needs to send images to a third-party cloud service.The achieved accuracy of pre-trained Inception v3 model is 99.99%.Therefore, the proposed mobile solution can serve as a reliable tool that can be used for melanoma detection by dermatologists and individual users.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".