PolyDiseaseNet: A Multimodal Machine Learning Approach for Comprehensive Disease Diagnosis and Prognosis
Bibliographic record
Abstract
Cancer is a leading cause of death world wide, accounting for an estimated 10 million deaths in 2020 alone. Early detection and accurate diagnosis are essential for improving patient outcomes and reducing cancer- related mortality. However, traditional diagnostic methods, such as biopsies and imaging, can be invasive, time- consuming, and expensive. Deep learning, a subfield of artificial intelligence, has emerged as a promising tool for cancer prediction and diagnosis. Deep learning algorithms can be trained on large datasets of gene expression data to identify complex patterns that are associated with diverse types of cancer. Once trained, these algorithms can be used to predict the type of cancer that a patient has based on their gene expression profile. Keras, a high- level Python library for TensorFlow, provides a user-friendly framework for building and training deep learning models. Keras offers a variety of pre-built deep learning architectures, including convolutional neural networks (CNNs) and recurrent neural networks (RNNs). CNNs are particularly wellsuited for image analysis tasks, while RNNs are well- suited for sequence data analysis tasks. Gene expression data is a rich source of information for identifying cancer-associated patterns. Gene expression data reflects the activity of genes within cells, and different genes are expressed at distinct levels in diverse types of cells and tissues. By analyzing gene expression patterns, deep learning algorithms can identify subtle differences between cancer cells and normal cells. This study investigated the application of deep learning using Keras for cancer prediction and diagnosis, with a primary focus on five prevalent cancer types: colon, breast, lung, kidney, and prostate. The study utilized gene expression data from a large cohort of patients to develop and evaluate deep learning models. Various deep learning architectures were explored, including CNNs and RNNs. The performance of each architecture was evaluated on a held-out test set. The bestperforming model achieved a classification accuracy of over 95% for each of the five cancer types. The findings of this study demonstrate the potential of deep learning using Keras to improve cancer prediction and diagnosis. The developed models achieved high classification accuracies, surpassing traditional methods. By analyzing gene expression data, deep learning models can identify specific genetic mutations or signaling pathways that are driving cancer growth and progression. This information can be used to select targeted therapies that are most likely to beeffective for each individual patient.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.008 | 0.003 |
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".