Suggested AI Tool Ideas: Complexity Ranking & Developer Prompts

Develop a single-page web application that accesses the user's webcam. Implement an existing open-source face tracking library to detect the eyes. Focus on calculating and visualizing **pupil dilation** changes over time and tracking the user's **gaze direction**. The output should be a simple numerical score or graph representing estimated mental effort.

Create a webcam application using a face mesh library to detect key facial landmarks. The system should capture a still frame and calculate a score based on the **symmetry of landmarks** (e.g., mouth corners, eye positions). The application should provide a quick visual and numerical assessment of facial asymmetry for rapid neurological screening.

Build a webcam application that isolates the user's extended hand. Use an object detection model (like a simplified YOLO) to track a specific hand region or marker with high temporal resolution. The core task is to analyze the **small, rapid changes in the hand's pixel position** over a 10-second period. Output should be an instability index, potentially using a simple **logistic regression or SVM** classification on the movement data to flag a "tremor" state.

Develop a sophisticated webcam tool that first uses **face detection** to locate the face. Implement **lighting normalization** techniques on the facial area. The core logic involves feeding the RGB data from the forehead and cheeks into a pre-trained **CNN or regression model** (running in the browser, e.g., via TensorFlow.js) that outputs an approximate relative skin temperature based on subtle color/brightness shifts.

Create a webcam tool focused on tracking the upper torso. Use an object tracking or pose estimation model (like YOLO) to identify and track the **rhythmic, subtle movements of the chest and shoulders**. The application must filter noise and accurately count the **cycles of movement** over a defined period to calculate and display the user's **respiration rate**.

This complex application requires the combination of multiple high-level components: 1) A face tracking model for **micro-expressions**. 2) A **PPG algorithm** (via webcam) for heart rate variability. 3) A separate library for real-time **voice tone/pitch analysis** (if feasible without a backend). The application must then fuse these three data streams (facial, circulatory, vocal) using a weighted model to provide a comprehensive, real-time **stress or anxiety score**.

Build a web application that can load and visualize DICOM medical imaging files directly in the browser using only vanilla JavaScript. The application should decode the image data and display relevant metadata, enabling client-side analysis without relying on external frameworks or libraries.

This is a conceptual application, likely requiring complex cellular automation or reaction-diffusion models. Prompt: Design a web application interface for a deterministic cellular simulation model that visualizes the spread and interaction of virtual "cancer cells" within a simulated tissue environment, allowing users to adjust parameters like nutrient levels and immune response.

Build a frontend application where a user can paste a raw DNA or RNA sequence. The application should parse the string and visually highlight key features such as start codons (AUG), stop codons (UAA, UAG, UGA), and user-defined motifs. The core logic involves efficient string searching and DOM manipulation to render the annotated sequence.

Create a web tool that renders a directed graph of a biological pathway (e.g., Glycolysis) from a structured data source like JSON or GraphML. Use a graph visualization library (like D3.js or Cytoscape.js) to draw nodes (metabolites) and edges (enzymes/reactions). The interface should allow users to click on nodes or edges to see more information.