add blacklists and change name

README.md

@@ -1,5 +1,5 @@
 ---
-title: PubMed AI Explorer
+title: Medical AI Wiki
 emoji: 🔬
 colorFrom: blue
 colorTo: purple
@@ -7,7 +7,7 @@ sdk: docker
 app_port: 3001
 ---
 
-# PubMed AI Explorer
+# Medical AI Wiki
 
 A web application that helps users explore AI algorithms used in medical research papers from PubMed.
 

backend/routes/search.js

@@ -50,27 +50,47 @@ async function searchAlgorithmUsage(problem, algorithmKey, algorithmData) {
       `("${problem}" AND "${synonym}")`
     ).join(' OR ');
     
+    // Build blacklist exclusions if they exist
+    let blacklistExclusions = '';
+    if (algorithmData.blacklist && algorithmData.blacklist.length > 0) {
+      const blacklistTerms = algorithmData.blacklist.map(term => `NOT "${term}"`).join(' ');
+      blacklistExclusions = ` ${blacklistTerms}`;
+    }
+    
     // Add filters to exclude review papers, meta-analyses, and systematic reviews
-    const filteredQuery = `(${synonymQueries}) NOT Review[Publication Type] NOT Meta-Analysis[Publication Type] NOT Systematic Review[Publication Type]`;
+    const filteredQuery = `(${synonymQueries})${blacklistExclusions} NOT Review[Publication Type] NOT Meta-Analysis[Publication Type] NOT Systematic Review[Publication Type]`;
     
     const searchUrl = `${PUBMED_BASE_URL}/esearch.fcgi?db=pubmed&term=${encodeURIComponent(filteredQuery)}&retmode=json`;
     
-    // Debug logging for CNN specifically
+    // Debug logging for CNN and GAN specifically
     if (algorithmKey === 'cnn') {
       console.log(`CNN Search for "${problem}":`);
       console.log(`Query: ${filteredQuery}`);
       console.log(`URL: ${searchUrl}`);
     }
     
+    if (algorithmKey === 'gan') {
+      console.log(`GAN Search for "${problem}":`);
+      console.log(`Synonym queries: ${synonymQueries}`);
+      console.log(`Blacklist exclusions: ${blacklistExclusions}`);
+      console.log(`Final query: ${filteredQuery}`);
+      console.log(`URL: ${searchUrl}`);
+    }
+    
     const response = await axios.get(searchUrl);
     const count = parseInt(response.data.esearchresult.count) || 0;
     
-    // Debug logging for CNN results
+    // Debug logging for CNN and GAN results
     if (algorithmKey === 'cnn') {
       console.log(`CNN Results: ${count} papers found`);
       console.log(`Sample IDs:`, response.data.esearchresult.idlist?.slice(0, 3));
     }
     
+    if (algorithmKey === 'gan') {
+      console.log(`GAN Results: ${count} papers found`);
+      console.log(`Sample IDs:`, response.data.esearchresult.idlist?.slice(0, 3));
+    }
+    
     return {
       algorithm: algorithmKey,
       name: algorithmData.name,
@@ -136,8 +156,15 @@ async function fetchAlgorithmCount(key, algo) {
 
   const generalQuery = algo.synonyms.map(s => `"${s}"`).join(' OR ');
   
+  // Build blacklist exclusions if they exist
+  let blacklistExclusions = '';
+  if (algo.blacklist && algo.blacklist.length > 0) {
+    const blacklistTerms = algo.blacklist.map(term => `NOT "${term}"`).join(' ');
+    blacklistExclusions = ` ${blacklistTerms}`;
+  }
+  
   // Add filters to exclude review papers, meta-analyses, and systematic reviews
-  const filteredQuery = `(${generalQuery}) NOT Review[Publication Type] NOT Meta-Analysis[Publication Type] NOT Systematic Review[Publication Type]`;
+  const filteredQuery = `(${generalQuery})${blacklistExclusions} NOT Review[Publication Type] NOT Meta-Analysis[Publication Type] NOT Systematic Review[Publication Type]`;
   
   try {
     const searchUrl = `${PUBMED_BASE_URL}/esearch.fcgi?db=pubmed&term=${encodeURIComponent(filteredQuery)}&retmode=json`;
@@ -203,8 +230,15 @@ router.get('/pubmed-link', (req, res) => {
     `("${problem}" AND "${synonym}")`
   ).join(' OR ');
   
+  // Build blacklist exclusions if they exist
+  let blacklistExclusions = '';
+  if (algoData.blacklist && algoData.blacklist.length > 0) {
+    const blacklistTerms = algoData.blacklist.map(term => `NOT "${term}"`).join(' ');
+    blacklistExclusions = ` ${blacklistTerms}`;
+  }
+  
   // Add filters to exclude review papers for PubMed links too
-  const filteredQuery = `(${synonymQueries}) NOT Review[Publication Type] NOT Meta-Analysis[Publication Type] NOT Systematic Review[Publication Type]`;
+  const filteredQuery = `(${synonymQueries})${blacklistExclusions} NOT Review[Publication Type] NOT Meta-Analysis[Publication Type] NOT Systematic Review[Publication Type]`;
   
   const pubmedUrl = `https://pubmed.ncbi.nlm.nih.gov/?term=${encodeURIComponent(filteredQuery)}`;
   
@@ -229,7 +263,15 @@ async function fetchAlgorithmCountByYear(key, algo, year, diskCache, retryCount
 
   const generalQuery = algo.synonyms.map(s => `"${s}"`).join(' OR ');
   const yearFilter = `"${year}"[Date - Publication]`;
-  const filteredQuery = `(${generalQuery}) AND ${yearFilter} NOT Review[Publication Type] NOT Meta-Analysis[Publication Type] NOT Systematic Review[Publication Type]`;
+  
+  // Build blacklist exclusions if they exist
+  let blacklistExclusions = '';
+  if (algo.blacklist && algo.blacklist.length > 0) {
+    const blacklistTerms = algo.blacklist.map(term => `NOT "${term}"`).join(' ');
+    blacklistExclusions = ` ${blacklistTerms}`;
+  }
+  
+  const filteredQuery = `(${generalQuery}) AND ${yearFilter}${blacklistExclusions} NOT Review[Publication Type] NOT Meta-Analysis[Publication Type] NOT Systematic Review[Publication Type]`;
   
   try {
     const searchUrl = `${PUBMED_BASE_URL}/esearch.fcgi?db=pubmed&term=${encodeURIComponent(filteredQuery)}&retmode=json`;

data/algorithms.json

@@ -4,235 +4,584 @@
       "name": "Support Vector Machine",
       "category": "classical_ml",
       "description": "A supervised learning algorithm that finds optimal hyperplanes for classification and regression tasks",
-      "synonyms": ["support vector machine", "SVM", "support vector classifier", "support vector regression", "SVR"]
+      "synonyms": [
+        "support vector machine",
+        "SVM",
+        "support vector classifier",
+        "support vector regression",
+        "SVR"
+      ],
+      "blacklist": [
+        "stroke volume monitoring",
+        "severe viral meningitis",
+        "syncope vasovagal mechanisms",
+        "superior vena cava",
+        "small vessel disease"
+      ]
     },
     "decision_tree": {
       "name": "Decision Tree",
-      "category": "classical_ml", 
+      "category": "classical_ml",
       "description": "A tree-like model that makes decisions by splitting data based on feature values",
-      "synonyms": ["decision tree", "decision trees", "DT", "CART", "classification tree", "regression tree"]
+      "synonyms": [
+        "decision tree",
+        "decision trees",
+        "DT",
+        "CART",
+        "classification tree",
+        "regression tree"
+      ]
     },
     "random_forest": {
       "name": "Random Forest",
       "category": "classical_ml",
       "description": "An ensemble method that combines multiple decision trees for improved accuracy",
-      "synonyms": ["random forest", "RF", "random forests", "forest classifier"]
+      "synonyms": [
+        "random forest",
+        "RF",
+        "random forests",
+        "forest classifier"
+      ],
+      "blacklist": [
+        "radiofrequency",
+        "rheumatoid factor",
+        "risk factor",
+        "renal failure",
+        "respiratory failure",
+        "reticular formation"
+      ]
     },
     "xgboost": {
       "name": "XGBoost",
       "category": "classical_ml",
       "description": "Extreme Gradient Boosting - an optimized gradient boosting framework",
-      "synonyms": ["xgboost", "XGBoost", "extreme gradient boosting", "XGB"]
+      "synonyms": [
+        "xgboost",
+        "XGBoost",
+        "extreme gradient boosting",
+        "XGB"
+      ]
     },
     "logistic_regression": {
       "name": "Logistic Regression",
       "category": "classical_ml",
       "description": "A linear model for binary and multiclass classification problems",
-      "synonyms": ["logistic regression", "logit", "logistic model", "LR"]
+      "synonyms": [
+        "logistic regression",
+        "logit",
+        "logistic model",
+        "LR"
+      ]
     },
     "naive_bayes": {
       "name": "Naive Bayes",
       "category": "classical_ml",
       "description": "A probabilistic classifier based on Bayes' theorem with independence assumptions",
-      "synonyms": ["naive bayes", "Naive Bayes", "NB", "Bayes classifier"]
+      "synonyms": [
+        "naive bayes",
+        "Naive Bayes",
+        "NB",
+        "Bayes classifier"
+      ]
     },
     "knn": {
       "name": "K-Nearest Neighbors",
       "category": "classical_ml",
       "description": "A non-parametric method that classifies data points based on the class of their nearest neighbors",
-      "synonyms": ["k-nearest neighbors", "KNN", "k-NN", "nearest neighbor", "k nearest neighbour"]
+      "synonyms": [
+        "k-nearest neighbors",
+        "KNN",
+        "k-NN",
+        "nearest neighbor",
+        "k nearest neighbour"
+      ]
     },
     "kmeans": {
       "name": "K-Means Clustering",
       "category": "classical_ml",
       "description": "An unsupervised clustering algorithm that partitions data into k clusters",
-      "synonyms": ["k-means", "K-means", "kmeans", "k-means clustering", "k means"]
+      "synonyms": [
+        "k-means",
+        "K-means",
+        "kmeans",
+        "k-means clustering",
+        "k means"
+      ]
     },
     "gradient_boosting": {
       "name": "Gradient Boosting",
       "category": "classical_ml",
       "description": "An ensemble method that builds models sequentially to correct errors of previous models",
-      "synonyms": ["gradient boosting", "GB", "GBM", "gradient boosted trees", "gradient boosting machine"]
+      "synonyms": [
+        "gradient boosting",
+        "GB",
+        "GBM",
+        "gradient boosted trees",
+        "gradient boosting machine"
+      ]
     },
     "ada_boost": {
       "name": "AdaBoost",
       "category": "classical_ml",
       "description": "Adaptive Boosting algorithm that combines weak learners into a strong classifier",
-      "synonyms": ["AdaBoost", "ada boost", "adaptive boosting", "adaboost"]
+      "synonyms": [
+        "AdaBoost",
+        "ada boost",
+        "adaptive boosting",
+        "adaboost"
+      ]
     },
     "pca": {
       "name": "Principal Component Analysis",
       "category": "classical_ml",
       "description": "A dimensionality reduction technique that finds principal components of data variance",
-      "synonyms": ["PCA", "principal component analysis", "principal components"]
+      "synonyms": [
+        "PCA",
+        "principal component analysis",
+        "principal components"
+      ],
+      "blacklist": [
+        "posterior cerebral artery",
+        "prostate cancer antigen",
+        "patient-controlled analgesia",
+        "percutaneous coronary angioplasty",
+        "primary care physician",
+        "polycystic ovary syndrome"
+      ]
     },
     "linear_regression": {
       "name": "Linear Regression",
       "category": "classical_ml",
       "description": "A linear approach to modeling the relationship between variables",
-      "synonyms": ["linear regression", "ordinary least squares", "OLS", "linear model"]
+      "synonyms": [
+        "linear regression",
+        "ordinary least squares",
+        "OLS",
+        "linear model"
+      ]
     },
     "cnn": {
       "name": "Convolutional Neural Network",
       "category": "deep_learning",
       "description": "Deep learning architecture specialized for processing grid-like data such as images",
-      "synonyms": ["convolutional neural network", "CNN", "ConvNet", "convolutional network", "deep convolutional", "conv neural network", "convolution neural network"]
+      "synonyms": [
+        "convolutional neural network",
+        "CNN",
+        "ConvNet",
+        "convolutional network",
+        "deep convolutional",
+        "conv neural network",
+        "convolution neural network"
+      ],
+      "blacklist": [
+        "cranial nerve nuclei",
+        "central nervous system",
+        "chronic kidney disease",
+        "clinical nurse navigator",
+        "calcineurin inhibitor"
+      ]
     },
     "lstm": {
       "name": "Long Short-Term Memory",
       "category": "deep_learning",
       "description": "A type of recurrent neural network capable of learning long-term dependencies",
-      "synonyms": ["LSTM", "long short-term memory", "long short term memory", "LSTM network"]
+      "synonyms": [
+        "LSTM",
+        "long short-term memory",
+        "long short term memory",
+        "LSTM network"
+      ]
     },
     "transformer": {
       "name": "Transformer",
       "category": "deep_learning",
       "description": "Attention-based neural network architecture for sequence-to-sequence tasks",
-      "synonyms": ["transformer", "transformers", "transformer model", "transformer architecture", "self-attention"]
+      "synonyms": [
+        "transformer",
+        "transformers",
+        "transformer model",
+        "transformer architecture",
+        "self-attention"
+      ]
     },
     "resnet": {
       "name": "ResNet",
       "category": "deep_learning",
       "description": "Residual Neural Network - a deep CNN architecture with skip connections",
-      "synonyms": ["ResNet", "resnet", "residual network", "residual neural network"]
+      "synonyms": [
+        "ResNet",
+        "resnet",
+        "residual network",
+        "residual neural network"
+      ]
     },
     "unet": {
       "name": "U-Net",
       "category": "deep_learning",
       "description": "A CNN architecture for biomedical image segmentation with encoder-decoder structure",
-      "synonyms": ["U-Net", "UNet", "u-net", "unet"]
+      "synonyms": [
+        "U-Net",
+        "UNet",
+        "u-net",
+        "unet"
+      ]
     },
     "gan": {
       "name": "Generative Adversarial Network",
       "category": "deep_learning",
       "description": "A framework where two neural networks compete to generate realistic data",
-      "synonyms": ["GAN", "generative adversarial network", "generative adversarial networks", "GANs"]
+      "synonyms": [
+        "generative adversarial network",
+        "generative adversarial networks",
+        "GANs",
+        "GAN model",
+        "GAN network",
+        "adversarial network",
+        "adversarial training"
+      ],
+      "blacklist": [
+        "giant axonal neuropathy",
+        "Giant Axonal Neuropathy",
+        "GAN neuropathy",
+        "axonal neuropathy",
+        "ganglion",
+        "ganglia",
+        "ganglioside",
+        "gangliosides",
+        "ganglionic",
+        "gangrene",
+        "gangrenous",
+        "Ganoderma",
+        "ganoderic",
+        "ganciclovir",
+        "gastric antral nodularity",
+        "gonadotropin-releasing hormone antagonist",
+        "glucosamine",
+        "galactosamine",
+        "N-acetylgalactosamine",
+        "GalNAc"
+      ]
     },
     "autoencoder": {
       "name": "Autoencoder",
       "category": "deep_learning",
       "description": "Neural networks that learn efficient representations by encoding and decoding data",
-      "synonyms": ["autoencoder", "auto-encoder", "autoencoders", "variational autoencoder", "VAE"]
+      "synonyms": [
+        "autoencoder",
+        "auto-encoder",
+        "autoencoders",
+        "variational autoencoder",
+        "VAE"
+      ]
     },
     "vgg": {
       "name": "VGG",
       "category": "deep_learning",
       "description": "Very Deep Convolutional Networks - a CNN architecture with small convolution filters",
-      "synonyms": ["VGG", "VGG-16", "VGG-19", "VGGNet"]
+      "synonyms": [
+        "VGG",
+        "VGG-16",
+        "VGG-19",
+        "VGGNet"
+      ]
     },
     "inception": {
       "name": "Inception",
       "category": "deep_learning",
       "description": "CNN architecture that uses inception modules for efficient computation",
-      "synonyms": ["Inception", "GoogLeNet", "Inception-v3", "Inception-v4", "inception network"]
+      "synonyms": [
+        "Inception",
+        "GoogLeNet",
+        "Inception-v3",
+        "Inception-v4",
+        "inception network"
+      ]
     },
     "rnn": {
       "name": "Recurrent Neural Network",
       "category": "deep_learning",
       "description": "Neural networks with memory that can process sequences of data",
-      "synonyms": ["RNN", "recurrent neural network", "recurrent network", "RNNs"]
+      "synonyms": [
+        "RNN",
+        "recurrent neural network",
+        "recurrent network",
+        "RNNs"
+      ],
+      "blacklist": [
+        "ribonuclease",
+        "registered nurse navigator",
+        "reactive nitrogen species"
+      ]
     },
     "gru": {
       "name": "Gated Recurrent Unit",
       "category": "deep_learning",
       "description": "Simplified variant of LSTM with fewer parameters and faster training",
-      "synonyms": ["GRU", "gated recurrent unit", "gated recurrent units", "GRUs"]
+      "synonyms": [
+        "GRU",
+        "gated recurrent unit",
+        "gated recurrent units",
+        "GRUs"
+      ]
     },
     "yolo": {
       "name": "YOLO",
       "category": "deep_learning",
       "description": "You Only Look Once - real-time object detection algorithm",
-      "synonyms": ["YOLO", "you only look once", "YOLOv3", "YOLOv4", "YOLOv5"]
+      "synonyms": [
+        "YOLO",
+        "you only look once",
+        "YOLOv3",
+        "YOLOv4",
+        "YOLOv5"
+      ]
     },
     "capsnet": {
       "name": "Capsule Network",
       "category": "deep_learning",
       "description": "Neural network architecture that uses capsules to better model hierarchical relationships",
-      "synonyms": ["CapsNet", "capsule network", "capsule networks", "dynamic routing"]
-    },
-    "attention": {
-      "name": "Attention Mechanism",
-      "category": "deep_learning",
-      "description": "Mechanism that allows models to focus on relevant parts of input sequences",
-      "synonyms": ["attention mechanism", "attention", "multi-head attention", "scaled dot-product attention"]
+      "synonyms": [
+        "CapsNet",
+        "capsule network",
+        "capsule networks",
+        "dynamic routing"
+      ]
     },
     "gpt": {
       "name": "GPT",
       "category": "llms",
       "description": "Generative Pre-trained Transformer - OpenAI's large language model family",
-      "synonyms": ["GPT", "gpt", "generative pre-trained transformer", "ChatGPT", "GPT-3", "GPT-4", "GPT-4o", "OpenAI GPT"]
+      "synonyms": [
+        "GPT",
+        "gpt",
+        "generative pre-trained transformer",
+        "ChatGPT",
+        "GPT-3",
+        "GPT-4",
+        "GPT-4o",
+        "OpenAI GPT"
+      ],
+      "blacklist": [
+        "glucose-6-phosphate transporter",
+        "glutamic pyruvic transaminase",
+        "glutathione peroxidase",
+        "glycerophosphate",
+        "guanosine triphosphate"
+      ]
     },
     "claude": {
       "name": "Claude",
       "category": "llms",
       "description": "Anthropic's AI assistant and large language model family",
-      "synonyms": ["Claude", "claude", "Anthropic Claude", "Claude-3", "Claude Sonnet", "Claude Haiku", "Claude Opus"]
+      "synonyms": [
+        "Claude",
+        "claude",
+        "Anthropic Claude",
+        "Claude-3",
+        "Claude Sonnet",
+        "Claude Haiku",
+        "Claude Opus"
+      ]
     },
     "bert": {
       "name": "BERT",
       "category": "llms",
       "description": "Bidirectional Encoder Representations from Transformers - Google's pre-trained language model",
-      "synonyms": ["BERT", "bert", "bidirectional encoder representations", "BERT model", "Google BERT"]
+      "synonyms": [
+        "BERT",
+        "bert",
+        "bidirectional encoder representations",
+        "BERT model",
+        "Google BERT"
+      ],
+      "blacklist": [
+        "behavioral emergency response team",
+        "biomedical emergency response team",
+        "blood-retinal barrier transport",
+        "bronchial epithelial cell"
+      ]
     },
     "gemini": {
       "name": "Gemini",
       "category": "llms",
       "description": "Google's multimodal large language model family",
-      "synonyms": ["Gemini", "gemini", "Google Gemini", "Gemini Pro", "Gemini Ultra", "Gemini Nano"]
+      "synonyms": [
+        "Gemini",
+        "gemini",
+        "Google Gemini",
+        "Gemini Pro",
+        "Gemini Ultra",
+        "Gemini Nano"
+      ]
     },
     "llama": {
       "name": "LLaMA",
       "category": "llms",
       "description": "Large Language Model Meta AI - Meta's open-source language model family",
-      "synonyms": ["LLaMA", "llama", "Llama", "Meta LLaMA", "Llama-2", "Llama 2", "Llama 3", "Code Llama"]
+      "synonyms": [
+        "LLaMA",
+        "llama",
+        "Llama",
+        "Meta LLaMA",
+        "Llama-2",
+        "Llama 2",
+        "Llama 3",
+        "Code Llama"
+      ]
     },
     "qwen": {
       "name": "Qwen",
       "category": "llms",
       "description": "Alibaba's large language model series with multilingual capabilities",
-      "synonyms": ["Qwen", "qwen", "Alibaba Qwen", "Qwen-7B", "Qwen-14B", "Qwen-72B", "Tongyi Qianwen"]
+      "synonyms": [
+        "Qwen",
+        "qwen",
+        "Alibaba Qwen",
+        "Qwen-7B",
+        "Qwen-14B",
+        "Qwen-72B",
+        "Tongyi Qianwen"
+      ]
     },
     "deepseek": {
       "name": "DeepSeek",
       "category": "llms",
       "description": "DeepSeek's large language model optimized for code and reasoning",
-      "synonyms": ["DeepSeek", "deepseek", "DeepSeek Coder", "DeepSeek LLM", "DeepSeek-V2"]
+      "synonyms": [
+        "DeepSeek",
+        "deepseek",
+        "DeepSeek Coder",
+        "DeepSeek LLM",
+        "DeepSeek-V2"
+      ]
     },
     "mistral": {
       "name": "Mistral",
       "category": "llms",
       "description": "Mistral AI's efficient and powerful open-source language models",
-      "synonyms": ["Mistral", "mistral", "Mistral 7B", "Mixtral", "Mistral AI", "Mixtral 8x7B"]
+      "synonyms": [
+        "Mistral",
+        "mistral",
+        "Mistral 7B",
+        "Mixtral",
+        "Mistral AI",
+        "Mixtral 8x7B"
+      ]
     },
     "palm": {
       "name": "PaLM",
       "category": "llms",
       "description": "Pathways Language Model - Google's large-scale language model",
-      "synonyms": ["PaLM", "palm", "Pathways Language Model", "PaLM-2", "Google PaLM"]
+      "synonyms": [
+        "PaLM",
+        "palm",
+        "Pathways Language Model",
+        "PaLM-2",
+        "Google PaLM"
+      ]
     },
     "t5": {
       "name": "T5",
       "category": "llms",
       "description": "Text-to-Text Transfer Transformer - Google's unified text processing model",
-      "synonyms": ["T5", "t5", "text-to-text transfer transformer", "Google T5"]
+      "synonyms": [
+        "T5",
+        "t5",
+        "text-to-text transfer transformer",
+        "Google T5"
+      ],
+      "blacklist": [
+        "T5 vertebra",
+        "T5 spinal",
+        "fifth thoracic vertebra",
+        "thoracic vertebra 5",
+        "T5 nerve root",
+        "T5 dermatome"
+      ]
     },
     "roberta": {
       "name": "RoBERTa",
       "category": "llms",
       "description": "Robustly Optimized BERT Pretraining Approach - Meta's improved BERT variant",
-      "synonyms": ["RoBERTa", "roberta", "robustly optimized BERT", "Meta RoBERTa"]
+      "synonyms": [
+        "RoBERTa",
+        "roberta",
+        "robustly optimized BERT",
+        "Meta RoBERTa"
+      ]
     },
     "phi": {
       "name": "Phi",
       "category": "llms",
       "description": "Microsoft's small language model series optimized for efficiency",
-      "synonyms": ["Phi", "phi", "Microsoft Phi", "Phi-3", "Phi-2", "Phi-1"]
+      "synonyms": [
+        "Microsoft Phi",
+        "Microsoft Phi-3",
+        "Microsoft Phi-2", 
+        "Microsoft Phi-1",
+        "Phi language model",
+        "Phi LLM",
+        "Phi small language model"
+      ],
+      "blacklist": [
+        "phi angle",
+        "phi coefficient",
+        "phi correlation",
+        "dihedral angle",
+        "phi psi angles",
+        "ramachandran plot",
+        "protein phi",
+        "phi torsion",
+        "golden ratio phi",
+        "phi statistic",
+        "phi phenomenon",
+        "phi value analysis",
+        "protected health information",
+        "personal health information",
+        "phosphatidylinositol",
+        "bacteriophage phi",
+        "phage phi",
+        "phi bacteriophage",
+        "phi X174",
+        "phi 6",
+        "phi 29",
+        "phi92",
+        "magnetic flux phi",
+        "flux phi",
+        "phi magnetic",
+        "volumetric flux",
+        "flow rate phi",
+        "phi value",
+        "phi values",
+        "phi analysis",
+        "phi-value analysis",
+        "protein folding phi",
+        "transition state phi",
+        "nucleation condensation phi",
+        "pharmacologic MRI phi",
+        "phMRI",
+        "DICOM phi",
+        "medical imaging phi",
+        "phi function",
+        "phi distribution",
+        "phi parameter",
+        "phi variable",
+        "phi measurement",
+        "phi calculation"
+      ]
     },
     "falcon": {
       "name": "Falcon",
       "category": "llms",
       "description": "Technology Innovation Institute's open-source large language model",
-      "synonyms": ["Falcon", "falcon", "Falcon-7B", "Falcon-40B", "Falcon-180B", "TII Falcon"]
+      "synonyms": [
+        "Falcon",
+        "falcon",
+        "Falcon-7B",
+        "Falcon-40B",
+        "Falcon-180B",
+        "TII Falcon"
+      ]
     }
   }
 }

data/timeline-cache.json

@@ -1,14 +1,14 @@
 {
-  "svm-2015": 2191,
-  "svm-2016": 2220,
-  "svm-2017": 2496,
-  "svm-2018": 2800,
-  "svm-2019": 3138,
+  "svm-2015": 2189,
+  "svm-2016": 2218,
+  "svm-2017": 2489,
+  "svm-2018": 2799,
+  "svm-2019": 3134,
   "svm-2020": 3552,
-  "svm-2021": 4402,
-  "svm-2022": 5312,
+  "svm-2021": 4400,
+  "svm-2022": 5310,
   "svm-2023": 5015,
-  "svm-2024": 5569,
+  "svm-2024": 5566,
   "decision_tree-2015": 2690,
   "decision_tree-2016": 2822,
   "decision_tree-2017": 3089,
@@ -17,18 +17,18 @@
   "decision_tree-2020": 4730,
   "decision_tree-2021": 5816,
   "decision_tree-2022": 6297,
-  "decision_tree-2023": 6106,
-  "decision_tree-2024": 6727,
-  "random_forest-2015": 2927,
-  "random_forest-2016": 3092,
-  "random_forest-2017": 3413,
-  "random_forest-2018": 4008,
-  "random_forest-2019": 4797,
-  "random_forest-2020": 6107,
-  "random_forest-2021": 8016,
-  "random_forest-2022": 9246,
-  "random_forest-2023": 9684,
-  "random_forest-2024": 11618,
+  "decision_tree-2023": 6107,
+  "decision_tree-2024": 6736,
+  "random_forest-2015": 2150,
+  "random_forest-2016": 2312,
+  "random_forest-2017": 2703,
+  "random_forest-2018": 3253,
+  "random_forest-2019": 4046,
+  "random_forest-2020": 5279,
+  "random_forest-2021": 7049,
+  "random_forest-2022": 8274,
+  "random_forest-2023": 8818,
+  "random_forest-2024": 10650,
   "xgboost-2015": 1,
   "xgboost-2016": 3,
   "xgboost-2017": 16,
@@ -41,14 +41,14 @@
   "xgboost-2024": 3269,
   "logistic_regression-2015": 24898,
   "logistic_regression-2016": 26330,
-  "logistic_regression-2017": 28299,
-  "logistic_regression-2018": 30524,
-  "logistic_regression-2019": 34055,
+  "logistic_regression-2017": 28288,
+  "logistic_regression-2018": 30522,
+  "logistic_regression-2019": 34054,
   "logistic_regression-2020": 41392,
   "logistic_regression-2021": 50426,
-  "logistic_regression-2022": 53795,
-  "logistic_regression-2023": 50528,
-  "logistic_regression-2024": 54933,
+  "logistic_regression-2022": 53799,
+  "logistic_regression-2023": 50537,
+  "logistic_regression-2024": 54946,
   "naive_bayes-2015": 1473,
   "naive_bayes-2016": 1641,
   "naive_bayes-2017": 1790,
@@ -58,7 +58,7 @@
   "naive_bayes-2021": 3505,
   "naive_bayes-2022": 3738,
   "naive_bayes-2023": 4030,
-  "naive_bayes-2024": 4923,
+  "naive_bayes-2024": 4924,
   "knn-2015": 574,
   "knn-2016": 604,
   "knn-2017": 595,
@@ -68,27 +68,27 @@
   "knn-2021": 1507,
   "knn-2022": 1829,
   "knn-2023": 1875,
-  "knn-2024": 2144,
+  "knn-2024": 2142,
   "kmeans-2015": 345,
   "kmeans-2016": 353,
-  "kmeans-2017": 424,
+  "kmeans-2017": 423,
   "kmeans-2018": 490,
   "kmeans-2019": 584,
   "kmeans-2020": 730,
-  "kmeans-2021": 1016,
+  "kmeans-2021": 1015,
   "kmeans-2022": 1255,
   "kmeans-2023": 1272,
-  "kmeans-2024": 1373,
+  "kmeans-2024": 1374,
   "gradient_boosting-2015": 2774,
   "gradient_boosting-2016": 2787,
-  "gradient_boosting-2017": 3104,
+  "gradient_boosting-2017": 3101,
   "gradient_boosting-2018": 3688,
   "gradient_boosting-2019": 4101,
   "gradient_boosting-2020": 4975,
   "gradient_boosting-2021": 6060,
   "gradient_boosting-2022": 6674,
   "gradient_boosting-2023": 6827,
-  "gradient_boosting-2024": 7834,
+  "gradient_boosting-2024": 7836,
   "ada_boost-2015": 67,
   "ada_boost-2016": 76,
   "ada_boost-2017": 74,
@@ -99,36 +99,36 @@
   "ada_boost-2022": 363,
   "ada_boost-2023": 370,
   "ada_boost-2024": 499,
-  "pca-2015": 6571,
-  "pca-2016": 6845,
-  "pca-2017": 7096,
-  "pca-2018": 7622,
-  "pca-2019": 8342,
-  "pca-2020": 9232,
-  "pca-2021": 10167,
-  "pca-2022": 10370,
-  "pca-2023": 9496,
-  "pca-2024": 9824,
+  "pca-2015": 6415,
+  "pca-2016": 6688,
+  "pca-2017": 6963,
+  "pca-2018": 7483,
+  "pca-2019": 8223,
+  "pca-2020": 9088,
+  "pca-2021": 9994,
+  "pca-2022": 10221,
+  "pca-2023": 9369,
+  "pca-2024": 9672,
   "linear_regression-2015": 9508,
   "linear_regression-2016": 10092,
-  "linear_regression-2017": 10640,
-  "linear_regression-2018": 11826,
-  "linear_regression-2019": 13019,
+  "linear_regression-2017": 10636,
+  "linear_regression-2018": 11825,
+  "linear_regression-2019": 13020,
   "linear_regression-2020": 15565,
-  "linear_regression-2021": 18227,
+  "linear_regression-2021": 18228,
   "linear_regression-2022": 19244,
-  "linear_regression-2023": 18409,
-  "linear_regression-2024": 19206,
+  "linear_regression-2023": 18411,
+  "linear_regression-2024": 19204,
   "cnn-2015": 116,
   "cnn-2016": 273,
-  "cnn-2017": 719,
-  "cnn-2018": 1676,
-  "cnn-2019": 2912,
-  "cnn-2020": 4256,
-  "cnn-2021": 6357,
-  "cnn-2022": 8084,
-  "cnn-2023": 7326,
-  "cnn-2024": 7451,
+  "cnn-2017": 718,
+  "cnn-2018": 1671,
+  "cnn-2019": 2905,
+  "cnn-2020": 4246,
+  "cnn-2021": 6342,
+  "cnn-2022": 8057,
+  "cnn-2023": 7299,
+  "cnn-2024": 7416,
   "lstm-2015": 21,
   "lstm-2016": 40,
   "lstm-2017": 117,
@@ -138,18 +138,17 @@
   "lstm-2021": 1334,
   "lstm-2022": 1969,
   "lstm-2023": 1826,
-  "lstm-2024": 2059,
+  "lstm-2024": 2060,
   "transformer-2015": 111,
   "transformer-2016": 108,
-  "transformer-2017": 102,
+  "transformer-2017": 101,
   "transformer-2018": 138,
   "transformer-2019": 184,
   "transformer-2020": 354,
   "transformer-2021": 683,
   "transformer-2022": 1698,
-  "transformer-2023": 3208,
-  "transformer-2024": 4961,
-  "resnet-2015": 2,
+  "transformer-2023": 3209,
+  "transformer-2024": 4962,
   "resnet-2016": 4,
   "resnet-2017": 23,
   "resnet-2018": 65,
@@ -167,28 +166,28 @@
   "unet-2020": 670,
   "unet-2021": 1103,
   "unet-2022": 1552,
-  "unet-2023": 1698,
-  "unet-2024": 1818,
-  "gan-2015": 3093,
-  "gan-2016": 3349,
-  "gan-2017": 3312,
-  "gan-2018": 3863,
-  "gan-2019": 4344,
-  "gan-2020": 5126,
-  "gan-2021": 5986,
-  "gan-2022": 6676,
-  "gan-2023": 6611,
-  "gan-2024": 6944,
+  "unet-2023": 1699,
+  "unet-2024": 1819,
+  "gan-2015": 63,
+  "gan-2016": 65,
+  "gan-2017": 66,
+  "gan-2018": 178,
+  "gan-2019": 329,
+  "gan-2020": 629,
+  "gan-2021": 979,
+  "gan-2022": 1237,
+  "gan-2023": 1267,
+  "gan-2024": 1255,
   "autoencoder-2015": 44,
   "autoencoder-2016": 102,
-  "autoencoder-2017": 162,
+  "autoencoder-2017": 161,
   "autoencoder-2018": 300,
   "autoencoder-2019": 397,
   "autoencoder-2020": 563,
   "autoencoder-2021": 821,
   "autoencoder-2022": 1075,
   "autoencoder-2023": 1181,
-  "autoencoder-2024": 1358,
+  "autoencoder-2024": 1359,
   "vgg-2015": 4,
   "vgg-2016": 10,
   "vgg-2017": 25,
@@ -207,8 +206,8 @@
   "inception-2020": 1757,
   "inception-2021": 1892,
   "inception-2022": 1995,
-  "inception-2023": 1931,
-  "inception-2024": 2033,
+  "inception-2023": 1930,
+  "inception-2024": 2026,
   "rnn-2015": 105,
   "rnn-2016": 116,
   "rnn-2017": 166,
@@ -238,7 +237,7 @@
   "yolo-2021": 246,
   "yolo-2022": 588,
   "yolo-2023": 774,
-  "yolo-2024": 886,
+  "yolo-2024": 887,
   "capsnet-2015": 8,
   "capsnet-2016": 4,
   "capsnet-2017": 5,
@@ -249,46 +248,36 @@
   "capsnet-2022": 103,
   "capsnet-2023": 82,
   "capsnet-2024": 73,
-  "attention-2015": 22456,
-  "attention-2016": 23429,
-  "attention-2017": 24706,
-  "attention-2018": 26810,
-  "attention-2019": 29325,
-  "attention-2020": 35137,
-  "attention-2021": 40508,
-  "attention-2022": 44763,
-  "attention-2023": 44196,
-  "attention-2024": 46082,
-  "gpt-2015": 110,
-  "gpt-2016": 93,
-  "gpt-2017": 96,
-  "gpt-2018": 96,
-  "gpt-2019": 97,
-  "gpt-2020": 118,
-  "gpt-2021": 151,
-  "gpt-2022": 167,
-  "gpt-2023": 2314,
-  "gpt-2024": 4243,
+  "gpt-2015": 88,
+  "gpt-2016": 69,
+  "gpt-2017": 74,
+  "gpt-2018": 75,
+  "gpt-2019": 81,
+  "gpt-2020": 96,
+  "gpt-2021": 126,
+  "gpt-2022": 144,
+  "gpt-2023": 2294,
+  "gpt-2024": 4217,
   "claude-2015": 1555,
   "claude-2016": 1864,
-  "claude-2017": 2127,
+  "claude-2017": 2125,
   "claude-2018": 2282,
   "claude-2019": 2379,
   "claude-2020": 2694,
   "claude-2021": 3013,
   "claude-2022": 2895,
   "claude-2023": 2760,
-  "claude-2024": 2979,
+  "claude-2024": 2980,
   "bert-2015": 147,
   "bert-2016": 152,
-  "bert-2017": 129,
+  "bert-2017": 128,
   "bert-2018": 142,
   "bert-2019": 141,
-  "bert-2020": 227,
+  "bert-2020": 226,
   "bert-2021": 391,
-  "bert-2022": 588,
-  "bert-2023": 577,
-  "bert-2024": 706,
+  "bert-2022": 587,
+  "bert-2023": 576,
+  "bert-2024": 704,
   "gemini-2015": 158,
   "gemini-2016": 181,
   "gemini-2017": 190,
@@ -308,7 +297,7 @@
   "llama-2021": 72,
   "llama-2022": 60,
   "llama-2023": 66,
-  "llama-2024": 224,
+  "llama-2024": 225,
   "qwen-2015": 0,
   "qwen-2016": 0,
   "qwen-2017": 0,
@@ -341,7 +330,7 @@
   "mistral-2024": 73,
   "palm-2015": 1078,
   "palm-2016": 1126,
-  "palm-2017": 1298,
+  "palm-2017": 1297,
   "palm-2018": 1297,
   "palm-2019": 1443,
   "palm-2020": 1685,
@@ -349,16 +338,16 @@
   "palm-2022": 1752,
   "palm-2023": 1708,
   "palm-2024": 1778,
-  "t5-2015": 554,
-  "t5-2016": 492,
-  "t5-2017": 601,
-  "t5-2018": 670,
-  "t5-2019": 699,
-  "t5-2020": 762,
-  "t5-2021": 850,
-  "t5-2022": 848,
-  "t5-2023": 890,
-  "t5-2024": 947,
+  "t5-2015": 524,
+  "t5-2016": 465,
+  "t5-2017": 583,
+  "t5-2018": 637,
+  "t5-2019": 661,
+  "t5-2020": 724,
+  "t5-2021": 807,
+  "t5-2022": 831,
+  "t5-2023": 857,
+  "t5-2024": 921,
   "roberta-2015": 144,
   "roberta-2016": 136,
   "roberta-2017": 140,
@@ -369,16 +358,16 @@
   "roberta-2022": 224,
   "roberta-2023": 267,
   "roberta-2024": 306,
-  "phi-2015": 758,
-  "phi-2016": 644,
-  "phi-2017": 812,
-  "phi-2018": 874,
-  "phi-2019": 937,
-  "phi-2020": 1028,
-  "phi-2021": 1094,
-  "phi-2022": 1167,
-  "phi-2023": 1172,
-  "phi-2024": 1230,
+  "phi-2015": 1,
+  "phi-2016": 1,
+  "phi-2017": 2,
+  "phi-2018": 5,
+  "phi-2019": 1,
+  "phi-2020": 2,
+  "phi-2021": 5,
+  "phi-2022": 5,
+  "phi-2023": 2,
+  "phi-2024": 6,
   "falcon-2015": 117,
   "falcon-2016": 122,
   "falcon-2017": 116,
@@ -388,5 +377,5 @@
   "falcon-2021": 215,
   "falcon-2022": 181,
   "falcon-2023": 184,
-  "falcon-2024": 182
+  "falcon-2024": 184
 }

frontend/index.html

@@ -1,13 +1,16 @@
 <!doctype html>
 <html lang="en">
-  <head>
-    <meta charset="UTF-8" />
-    <link rel="icon" type="image/svg+xml" href="/vite.svg" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <title>PubMed AI Explorer</title>
-  </head>
-  <body>
-    <div id="root"></div>
-    <script type="module" src="/src/main.tsx"></script>
-  </body>
+
+<head>
+  <meta charset="UTF-8" />
+  <link rel="icon" type="image/svg+xml" href="/vite.svg" />
+  <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+  <title>MedicalAIWiki</title>
+</head>
+
+<body>
+  <div id="root"></div>
+  <script type="module" src="/src/main.tsx"></script>
+</body>
+
 </html>

frontend/src/components/Navbar.tsx

@@ -12,7 +12,7 @@ const Navbar: React.FC = () => {
           <div className="flex items-center">
             <Link to="/" className="flex items-center space-x-2">
               <BarChart3 className="h-8 w-8 text-blue-600" />
-              <span className="text-xl font-bold text-gray-900">PubMed AI Explorer</span>
+              <span className="text-xl font-bold text-gray-900">Medical AI Wiki</span>
             </Link>
           </div>