{"id":1575,"date":"2026-01-25T15:18:48","date_gmt":"2026-01-25T21:18:48","guid":{"rendered":"https:\/\/zogen.local\/?p=1575"},"modified":"2026-01-25T15:18:48","modified_gmt":"2026-01-25T21:18:48","slug":"cancer-diagnosis-with-artificial-intelligence","status":"publish","type":"post","link":"https:\/\/zogen.mx\/en\/el-diagnostico-del-cancer-con-inteligencia-artificial\/","title":{"rendered":"Cancer diagnosis with artificial intelligence"},"content":{"rendered":"

He<\/strong> Cancer diagnosis with artificial intelligence and genomic testing<\/strong> It's a concept that has moved from theory to practice thanks to recent research that is revolutionizing oncology. One of the most significant advances is the TOAD artificial intelligence model, developed by MIT, which allows for highly accurate prediction of the origin of tumors that are difficult to identify. This technology, combined with genomic tests like those offered by Tempus, is ushering in a new era in personalized cancer diagnosis and treatment.<\/p>\n\n\n\n

<\/p>\n\n\n\n

The MIT study: How was TOAD developed?<\/h2>\n\n\n\n

In 2021, a group of researchers from the Massachusetts Institute of Technology (MIT) published in Nature Medicine<\/em> a study that presented the TOAD model (Tumor Origin Assessment via Deep learning<\/em>This system uses deep neural networks to analyze histological images\u2014photographs of tissue taken from biopsies\u2014with the aim of identifying the primary site of cancer, even when this is not evident to human pathologists.<\/p>\n\n\n\n

The study trained TOAD using over 20,000 histological samples from patients diagnosed with various types of cancer. Through machine learning, the model was able to recognize subtle microscopic patterns that differentiate one type of tumor from another. In tests with real clinical cases, TOAD was able to predict the tumor's origin with an accuracy of 83%, a figure significantly higher than the average for conventional diagnosis in cases of cancer of unknown origin.<\/p>\n\n\n\n

This precision can have a direct impact on treatment, since identifying the origin of the cancer allows for more effective therapies and avoids unnecessary interventions.<\/p>\n\n\n\n

<\/p>\n\n\n\n

What is cancer of unknown origin and why is it a challenge?<\/h2>\n\n\n\n

Cancer of unknown primary (CUP) accounts for between 21% and 51% of all cancer diagnoses. In these cases, although metastases are detected throughout the body, the exact location of the tumor's origin cannot be determined. This complicates treatment, as many therapies are designed specifically for certain types of cancer.<\/p>\n\n\n\n

The traditional process for trying to find the cause involves multiple tests, biopsies, CT scans, and empirical treatments. This not only delays appropriate care but also takes a toll on the patient emotionally and physically. This is where technologies like TOAD offer an innovative and precise alternative.<\/p>\n\n\n\n

<\/p>\n\n\n\n

What are genomic tests and how does Tempus help?<\/h2>\n\n\n\n

Genomic tests analyze tumor DNA to identify mutations and genetic alterations that may influence its behavior. These tests not only help detect the type of cancer but also predict its response to certain treatments.<\/p>\n\n\n\n

Tempus is one of the leading platforms in this field. Its technology allows for the sequencing of hundreds of genes from a tissue or blood sample, and then comparing the results with a massive database that includes clinical, molecular, and treatment response information from thousands of patients.<\/p>\n\n\n\n

This allows not only confirmation of the tumor's origin but also the suggestion of targeted therapies that have shown efficacy in similar cases. Furthermore, in some cases, it can avoid treatments like chemotherapy when the genetic profile indicates they would be ineffective.<\/p>\n\n\n\n

<\/p>\n\n\n\n

How do TOAD and Tempus complement each other?<\/h2>\n\n\n\n

The combined use of TOAD and Tempus exemplifies the power of integrating artificial intelligence with genomics. TOAD offers rapid prediction based on microscopic images; Tempus validates that prediction at the molecular level. This combination can:<\/p>\n\n\n\n