Lecture 13: Artificial intelligence applications in medicine and genomics - Aug 29 - 8:30 MX 6:30 DE

Lecture 13: Artificial intelligence applications in medicine and genomics - Aug 29 - 8:30 MX 6:30 DE

by Dra. Alejandra Cervera

Life & Natural Systems AI — News & Applications UYComputer scienceUYQArtificial intelligence (AI)
🎙 Dra. Alejandra Cervera 👥 4K 📅 August 30, 2025 ⏱ 51 min 👁 141 📄 expert opinion 🧭 2026-08-13
Available in: English (current) Français

Keywords

generative AIpredictive AIvariant callinggene expressionCRLF2

Summary

The lecture by Dr. Alejandra Cervera, deputy director of population genomics at INMEGEN, provides an overview of AI applications in medicine and genomics, with a focus on both generative and predictive AI. It begins by distinguishing these two types: generative AI (e.g., large language models) is used for summarizing medical information, aiding in clinical documentation, and supporting drug interaction checks, with examples like ClinicalKey AI. Predictive AI, based on classical machine learning, is applied in diagnostics (e.g., AI-assisted colonoscopy and mammography) and treatment direction (e.g., sepsis prediction and patient deterioration alerts), with evidence of reduced mortality. The lecture then shifts to genomics, explaining basic molecular biology and sequencing. In genomics, generative AI is emerging for sequence analysis, while predictive AI is used for variant calling and gene expression classification, exemplified by the MammaPrint test for breast cancer. Dr. Cervera presents her own project on B-cell acute lymphoblastic leukemia in Mexican children, which uses RNA sequencing and machine learning (k-means clustering) to detect CRLF2 overexpression and fusion genes, aiming to improve prognosis and treatment. The project operates as a clinical service, providing results to oncologists within days. Challenges include limited sample size and the need to expand classification to other subtypes. The lecture concludes with a Q&A about improving genomics in Mexico, emphasizing the need for funding and infrastructure.

218 words

Critical Evaluation

Value of the Information & Strength of the Argument

The lecture provides valuable information by bridging theoretical AI concepts with concrete clinical and genomic applications. It effectively argues for the utility of both generative and predictive AI, supporting claims with examples of FDA-approved tools and meta-analyses. The presentation of the speaker’s own research adds original insight, demonstrating a practical application of machine learning in a real-world medical context. The argumentation is solid, though some points could benefit from more detailed evidence or citations.

Scientific Rigor, Source Quality, Title Accuracy

The scientific rigor is generally high: the speaker is an expert, and the content aligns with current literature. Sources mentioned include a meta-analysis on AI-assisted colonoscopy and the MammaPrint test, but specific citations are not provided in the video. The title accurately reflects the content, and the lecture is well-structured. The description includes only the speaker’s name and the summer school context, with no additional links. The lecture is a single presentation, so no public comments are available for analysis.

169 words

Title / Content Match

The title accurately reflects the content: the lecture covers AI applications in medicine and genomics, with a focus on predictive and generative AI, and includes a case study in leukemia genomics.

Quality & Reliability

8/10

The lecture is given by a domain expert (deputy director of population genomics at INMEGEN) and presents both established clinical AI applications (with references to FDA-approved tools and meta-analyses) and original research from her own project. The content is well-structured, clearly distinguishes between generative and predictive AI, and provides concrete examples. However, some claims lack specific citations, and the presentation is a lecture rather than a peer-reviewed synthesis.

Key Moments

Markers derived by PSI from the transcript: the creator did not define chapters.

Cited Sources

  • ClinicalKey AI — Mentioned as a generative AI tool for medical information summarization
  • MammaPrint — Mentioned as a gene expression-based test for breast cancer prognosis
  • Meta-analysis on AI-assisted colonoscopy — Referenced as evidence for improved polyp detection

Concurring Sources

  • FDA approval of AI mammography — Mentioned as an example of AI in clinical practice

Contribution & Novelties

The lecture provides a comprehensive overview of AI in medicine and genomics, but its main novelty lies in the presentation of the speaker’s ongoing research on childhood leukemia in Mexico. This project applies machine learning to RNA-seq data to identify CRLF2 overexpression and fusion genes, which are associated with poor prognosis. The work is notable for its operational integration into clinical practice, providing rapid results to oncologists. The lecture also highlights the challenges of applying AI in resource-limited settings and the need for population-specific genomic data.

Pour aller plus loin :

  • CRLF2 gene — GeneCards entry for CRLF2, relevant to the discussed gene.
  • RNA sequencing — Wikipedia article on RNA-Seq, the technology used in the project.
  • K-means clustering — Wikipedia article on k-means, the algorithm used for expression clustering.

129 words

Radar Profile

The radar profile shows high scores in information quantity, quality, and reliability, with a slightly lower technical level. This indicates a well-balanced lecture that is both informative and credible, though it may require some background knowledge to fully appreciate the technical aspects.

Reliability 8/10