AI-Powered CNN Model for Automated Lung Cancer Diagnosis in Medical Imaging
Bibliographic record
Abstract
Lung cancer remains a critical health concern in the entire world, which has been a major cause of high rates of cancer-related mortalities that affect individuals in every part of the world. The findings emphasize the notable potential of deep learning procedures to assist radiologists in diagnosing cases of lung-related abnormalities appropriately. Such methods are also leading to the improvement of AI-based healthcare products. The enhancements to the suggested model [16, 17, 18, 21] in the future will be aimed at tuning hyperparameters, 3D CNN [16, 17, 18] architectures, and the integration of patient clinical data, with the aim of further increasing the accuracy [16, 17, 19] of diagnosis as well as system performance. This paper uses the IQ-OTHNCCD dataset, a publicly available and highly annotated set of CT imaging that has been annotated by experts in the medical field. The preprocessing techniques applied will involve changing the images to Grayscale, normalizing the pixel values, ensuring consistency in the images, and converting them to a standard size of 128x128 pixels, which is the ideal size to feed the images into the CNN [16, 17, 18]. In the proposed work, the model [16, 17, 18, 21] integrates multi-scale convolutional layers with adaptive dropout (rate=0.5) and ReLU activations, yielding 95% accuracy [16, 17, 19] and 0.95 F1-score (95% CI: 93.8–96.2%) on a 70/15/15 train/validation/test split— a 4% improvement in F1-score. Preprocessing includes grayscale conversion, pixel normalization to [0,1], and resizing to 128x128 pixels. The architecture comprises three convolutional blocks (32/64/128 filters, 3x3 kernels), max-pooling (2x2), flattening, a 512-unit dense layer, and a 3-unit softmax output. Future enhancements include hyperparameter tuning, 3D CNN [16, 17, 18] integration, and clinical data fusion to exceed 97% accuracy [16, 17, 19].
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
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".