Dreamento Dev Path

EEG, Sleep Staging & Lucid Dreams — A Programmer’s Roadmap

Lessons

  1. Before You Begin (free preview): Who this course is for, where it leads (real Dreamento development), and how the Learn/Do/Done-when structure works.
  2. Chapter 1 — EEG Basics (free preview): What EEG actually measures, the 10-20 system, and learning to spot artifacts (blinks, muscle noise) by eye — with a real hands-on Colab lab.
  3. Game — Artifact Spotter (free preview): Drag-and-drop game: sort EEG squiggle cards into Brain Signal, Eye Blink, Muscle Tension, or Electrical Noise.
  4. Chapter 2 — Sleep Architecture & AASM Scoring (free preview): The sleep stages (N1/N2/N3/REM), what defines each one visually, a drag-and-drop game, and a real Colab lab staging actual labelled epochs.
  5. Game — Sleep Stage Detective (free preview): Drag-and-drop game: match EEG waveform cards to the correct sleep stage floor of the building.
  6. Chapter 3 — The Frequency-Domain View (free preview): The EEG frequency bands (delta through beta), how to read a spectrogram, and a Colab lab computing one across a full real night.
  7. Game — Frequency Band Match (free preview): Drag-and-drop game: match Delta through Beta to their characteristic waveform speed.
  8. Chapter 4 — A Classical Baseline (free preview): Random Forest on band-power features, why subject-level splits matter, and reading a confusion matrix — the thing a real neural network has to beat.
  9. Game — Leaky or Legit? (free preview): Drag-and-drop game: sort train/test split scenarios into Legit or Leaky (data leakage).
  10. Chapter 5 — Deep Learning Prerequisites (free preview): CNNs and LSTMs explained from scratch with real analogies — what a 1D convolution actually does, and why memory across epochs matters.
  11. Game — Predict the Next Stage (free preview): Drag-and-drop game building intuition for why LSTMs use memory — predict the next sleep stage from a history of recent stages.
  12. Chapter 6 — Reimplement TinySleepNet (free preview): Combine the CNN and LSTM workshops into one real, trainable model — and discover a real TinySleepNet already sitting inside Dreamento’s own code.
  13. Chapter 7 — Read Critically: DeepSleepNet & U-Sleep (free preview): Two real architectures, two different trade-offs — two-lens CNN + bidirectional LSTM vs. a fully-convolutional design with no memory at all, and why neither is naturally built for live use.
  14. Chapter 8 — Connect It Back to Dreamento (free preview): The real model.py, nn.py, and realTimeAutoScoring.py code, the exact config Dreamento’s live scorer expects, and a concrete written plan for where a trained model would plug in.
  15. Chapter A1 — The Eye-Signaling Technique (free preview): How every "lucid" label in this module actually gets created — the LRLR signal, since the 1970s, plus a hands-on trace-recognition game.
  16. Chapter A2 — The "Hybrid State" Picture (free preview): Voss 2009: lucid REM as a mix of waking and dreaming EEG, frontal gamma, and why this is a first sketch rather than a finished map.
  17. Chapter A3 — The Gamma Hypothesis, Tested Causally (free preview): Voss 2014: 40Hz tACS inducing lucidity, and why this connects directly to Dreamento’s own stimulation features.
  18. Chapter A4 — Where the Field Stands Now (free preview): The 2025 Demirel et al. study (with Dresler as co-author) — the largest lucid dreaming EEG sample yet, and a real update to the 2009 picture.
  19. Chapter A5 — Connecting to Dreamento’s Real-Time Features (free preview): The real stimulate() function, the full real sensor list, and what to trace yourself in the repository for stimulation timing and marker logging.
  20. Chapter 9 — Where This Goes Next: Making It Usable for Everyone (free preview): A design proposal, not a build: a simple Flutter app backed by a server-side data API, so citizen scientists get a button and an honest answer instead of an Anaconda install.