Practical LLMs for Healthcare : Build generative AI models for wearable health signals using Python and Transformers

  • 予約

Practical LLMs for Healthcare : Build generative AI models for wearable health signals using Python and Transformers

  • ただいまウェブストアではご注文を受け付けておりません。 ⇒古書を探す
  • 製本 Paperback:紙装版/ペーパーバック版
  • 言語 ENG
  • 商品コード 9781807609634
  • DDC分類 006.31

Full Description

Build wearable health AI with Python: process PPG time series, train Transformer and diffusion models, run streaming inference, and ground explanations with RAG and medical knowledge.
Free with your book: DRM-free PDF version + access to Packt's next-gen Reader*

Key Features

Apply Transformers and diffusion models to real-world PPG signals from wearable devices
Build denoising, self-supervised learning, and streaming inference pipelines in Python
Connect wearable signals to medical knowledge with RAG, vector search, and grounded explanations

Book DescriptionFrom raw pulse signals to evidence-grounded AI explanations, learn to build the complete workflow. Wearable devices generate continuous photoplethysmography (PPG) data, but motion artifacts, missing samples, limited labels, and individual variability make it difficult to model. Practical LLMs for Healthcare shows you how to apply generative AI beyond text to the challenges of wearable biosignals. See how sequence-modeling ideas from LLMs can be adapted to physiological time series while accounting for noise, timing, and variability.
Using Python and real PPG datasets, you will visualize, clean, and tokenize signals; build denoising and Transformer-based time-series models; and pretrain on unlabeled data with self-supervised learning. Generate synthetic PPG with diffusion models, then develop streaming inference, alerting, and personalization workflows for wearable monitoring.
Next, connect PPG embeddings to medical knowledge with retrieval-augmented generation (RAG), FAISS vector search, biomedical text encoders, and structured prompts. Explore privacy safeguards, hallucination detection, faithfulness, and clinical benchmarking to evaluate the quality of AI-generated explanations. By the end, you'll understand how to build and assess end-to-end generative AI pipelines for wearable health applications.
*Email sign-up and proof of purchase requiredWhat you will learn

Preprocess and tokenize PPG time series for Transformer models
Build denoising models for noisy wearable health signals
Train self-supervised foundation models on unlabeled PPG data
Generate synthetic PPG signals using diffusion models
Run streaming inference for real-time health monitoring
Personalize models for individual physiological patterns
Connect PPG embeddings to medical knowledge using RAG and FAISS
Evaluate generated explanations for privacy, faithfulness, and hallucinations

Who this book is forThis book is for machine learning engineers, AI engineers, data scientists, health researchers, and software developers working with wearable sensors or health-tech applications who want to apply modern generative AI beyond text and images. Readers should be comfortable with Python, NumPy, and basic neural network concepts (experience with Keras or PyTorch is helpful). No prior knowledge of biology, signal processing, or healthcare is required.

Contents

Table of Contents

Foundations of Large Language Models for Healthcare
Understanding What's Inside a Large Language Model
Working with Raw PPG Data in Python
Tokenizing Continuous Signals for Transformer Models
Building Your First Generative Denoising Model
Transformer Architectures for Time-Series Data
Self-Supervised Learning on Unlabeled PPG Data
Synthetic PPG Data Generation with Diffusion Models
Streaming Inference for Real-Time PPG Monitoring
Building Alerting and Event-Detection Logic
Personalizing Models for Individual Users
Evaluating and Validating Generative Health Models
Multimodal Fusion: Combining PPG with Other Sensors
RAG for Health: Connecting Wearable Data to Medical Knowledge

最近チェックした商品