Last changes

.github/workflows/frontend_push.yml

@@ -13,6 +13,7 @@ env:
   ECS_SERVICE: medical_assistant_frontend
   ECR_REPOSITORY: 891377268337.dkr.ecr.us-east-1.amazonaws.com/medical_assistant_frontend
   API_KEY: ${{secrets.API_KEY}}
+  VITE_BACKEND_URL: http://lb-medical-assistant-503996398.us-east-1.elb.amazonaws.com:8000/
 
 jobs:
   build_image_frontend:
@@ -39,6 +40,7 @@ jobs:
           IMAGE_TAG: lastest
         run: |
 
+          echo VITE_BACKEND_URL=${{env.VITE_BACKEND_URL}} >> ./frontend/.env
           docker build -t $ECR_REPOSITORY:$IMAGE_TAG ./frontend
 
           docker push $ECR_REPOSITORY:$IMAGE_TAG

backend/api/v1/post_suggestions.py

@@ -4,7 +4,10 @@
 
 from controllers.v1.history_clinic import extract_text_from_pdf
 from controllers.v1.MRI_image import predict_tumor_by_img
-from controllers.v1.validate_files import processing_file_type_based
+from controllers.v1.validate_files import (
+    check_files,
+    type_based_checking,
+)
 
 from dotenv import load_dotenv
 from model import GenerativeModel
@@ -34,24 +37,22 @@
 
 
 @router.post(
-    "/files",
+    "/uploadfiles",
     status_code=200,
-    dependencies=[
-        Depends(processing_file_type_based),
-    ],
     response_model=dict[str, str],
+    dependencies=[Depends(check_files)],
 )
 async def predict_tumor(
     files: list[UploadFile],
 ):
-    img = None
-    pdf = None
-
-    for file in files:
-        if file.filename.split(".")[0] == "brain_mri":
-            img = file
-        elif file.filename.split(".")[0] == "clinical_history":
-            pdf = file
+    # Check the type of the files pydantic model based
+    dict_files = {
+        file.filename.split(".")[1]: type_based_checking(file) for file in files
+    }
+
+    # Get the image and the pdf
+    img = dict_files.get("jpeg") or dict_files.get("png") or dict_files.get("jpg")
+    pdf = dict_files.get("pdf")
 
     # Call the function to predict the tumor
     classification = await predict_tumor_by_img(img=img)

backend/controllers/v1/validate_files.py

@@ -1,7 +1,6 @@
 # Check if the number of files is correct
 from fastapi import HTTPException, UploadFile
 from pydantic import ValidationError
-import os
 
 from schemas.ClinicHistoryFileUploaded import ClinicHistoryFileUploaded
 from schemas.MRIFileUploaded import MRIFileUploaded
@@ -11,6 +10,17 @@ def validate_quantity_files(files: list):
     return len(files) != 2
 
 
+def check_files(files: list[UploadFile]):
+    # Check if each file has an extension
+    if not all(["." in file.filename for file in files]):
+        raise HTTPException(status_code=400, detail="File without extension.")
+
+    if validate_quantity_files(files):
+        raise HTTPException(
+            status_code=400, detail="The number of files is not correct."
+        )
+
+
 def is_type_PDF(type_file):
     return type_file == "application/pdf"
 
@@ -19,40 +29,25 @@ def is_type_image(type_file):
     return type_file in ["image/jpeg", "image/png", "image/jpg"]
 
 
-def processing_file_type_based(files: list[UploadFile]):
-    if validate_quantity_files(files):
+def type_based_checking(file: MRIFileUploaded | ClinicHistoryFileUploaded):
+    # Check the model type based on defined schema
+    if is_type_image(file.content_type):
+        try:
+            MRIFileUploaded(file=file, type=file.content_type, size=file.size)
+
+        except ValidationError as e:
+            raise HTTPException(status_code=400, detail=str(e).splitlines()[2])
+
+    elif is_type_PDF(file.content_type):
+        try:
+            ClinicHistoryFileUploaded(file=file, type=file.content_type, size=file.size)
+
+        except ValidationError as e:
+            raise HTTPException(status_code=400, detail=str(e).splitlines()[2])
+
+    else:
         raise HTTPException(
-            status_code=400, detail="The number of files is not correct."
+            status_code=400,
+            detail=f"The type of the file {file.filename} is not supported.",
         )
-
-    img = None
-    pdf = None
-
-    for file in files:
-        # Assign the file to the corresponding variable
-        if is_type_image(file.content_type):
-            try:
-                name = os.path.splitext(file.filename)[0]
-                img = MRIFileUploaded(
-                    file=file, name=name, type=file.content_type, size=file.size
-                )
-
-            except ValidationError as e:
-                raise HTTPException(status_code=400, detail=str(e).splitlines()[2])
-
-        elif is_type_PDF(file.content_type):
-            try:
-                name = os.path.splitext(file.filename)[0]
-                pdf = ClinicHistoryFileUploaded(
-                    file=file, name=name, type=file.content_type, size=file.size
-                )
-
-            except ValidationError as e:
-                raise HTTPException(status_code=400, detail=str(e).splitlines()[2])
-
-        else:
-            raise HTTPException(
-                status_code=400,
-                detail=f"The type of the file {file.filename} is not supported.",
-            )
-    return [img, pdf]
+    return file

backend/schemas/ClinicHistoryFileUploaded.py

@@ -5,7 +5,6 @@
 
 
 class ClinicHistoryFileUploaded(BaseModel):
-    name: Literal["clinical_history"]
     file: UploadFile
     type: Literal["application/pdf"]
     size: Annotated[int, field_validator("size")]

backend/schemas/MRIFileUploaded.py

@@ -5,7 +5,6 @@
 
 
 class MRIFileUploaded(BaseModel):
-    name: Literal["brain_mri"]
     file: UploadFile
     type: Literal["image/jpeg", "image/png", "image/jpg"]
     size: int

frontend/src/App.jsx

@@ -34,7 +34,9 @@ function App() {
   // The UploadFileWrapper component is used to render the form and the model's response data
   return (
     <div className="flex flex-col min-h-screen">
-      <h1 className="text-6xl font-bold p-6">Asistente Medico</h1>
+      <h1 className="text-6xl font-bold p-6">
+        Asistente de diagnóstico médico
+      </h1>
       <Notification notification={notification} />
       <main className="flex-1 flex max-w-7xl self-center pt-16 flex-col">
         <UploadFileWrapper

frontend/src/components/UploadFileForm.jsx

@@ -1,63 +1,66 @@
 import React from "react";
 
-function UploadFileForm({
-  handleFileUpload,
-  expectedFiles = ["clinical_history.pdf", "brain_mri.jpeg"],
-  notify,
-  disabled,
-}) {
+function UploadFileForm({ handleFileUpload, notify, disabled }) {
   const submitHandler = async (e) => {
     e.preventDefault();
-    const files = e.target.elements.file.files;
 
-    if (files.length === 0) {
-      notify("No hay archivos seleccionados", "error");
-      return;
-    }
+    const pdfFiles = e.target.elements.pdf.files;
+    const imgFiles = e.target.elements.img.files;
 
-    if (files.length !== expectedFiles.length) {
-      notify(
-        `Por favor, suba exactamente ${expectedFiles.length} archivos`,
-        "error",
-      );
+    if (pdfFiles.length === 0 && imgFiles.length === 0) {
+      notify("No ha seleccionado ningun archivo", "error");
       return;
-    }
-
-    const filenames = Array.from(files).map((file) => file.name);
-
-    if (
-      !expectedFiles.every((expectedFile) => filenames.includes(expectedFile))
-    ) {
-      notify("Los archivos no tienen el formato correcto", "error");
+    } else if (pdfFiles.length === 0) {
+      notify("No ha seleccionado la historia clinica", "error");
+      return;
+    } else if (imgFiles.length === 0) {
+      notify("No ha seleccionado la IRM (cerebro)", "error");
       return;
     }
 
     const formData = new FormData();
 
-    for (let i = 0; i < expectedFiles.length; i++) {
-      formData.append("files", files[i]);
+    for (let i = 0; i < pdfFiles.length; i++) {
+      formData.append("files", pdfFiles[i]);
+    }
+
+    for (let i = 0; i < imgFiles.length; i++) {
+      formData.append("files", imgFiles[i]);
     }
 
     handleFileUpload(formData);
-    notify("Archivos enviados exitosamente. Procesando...");
+    notify(
+      "Archivos enviados exitosamente. Por favor no refresque la pagina...",
+    );
   };
-
+  // la IRM (cerebro)
   return (
     <div className="pr-8">
       <form encType="multipart/form-data" onSubmit={submitHandler}>
-        <label className="text-2xl font-semibold" htmlFor="file">
-          Seleccione la IRM (cerebro) e historia clínica:
-        </label>
-        <p className="text-sm text-gray-500">
-          Formato esperado: clinical_history.pdf, brain_mri.jpeg
-        </p>
         <div className="flex flex-col ">
+          <label className="text-2xl font-semibold" htmlFor="pdf">
+            Seleccione la historia clínica
+          </label>
+          <p className="text-sm text-gray-500">
+            Solo se aceptan pdf con tamaño inferior a 2MB
+          </p>
+          <input
+            type="file"
+            className="file-input file-input-bordered w-full max-w-xs my-4"
+            id="pdf"
+            name="pdf"
+          />
+          <label className="text-2xl font-semibold" htmlFor="img">
+            Seleccione la IRM (cerebro)
+          </label>
+          <p className="text-sm text-gray-500">
+            Solo se aceptan imágenes con tamaño inferior a 2MB
+          </p>
           <input
             type="file"
             className="file-input file-input-bordered w-full max-w-xs my-4"
-            id="file"
-            name="file"
-            multiple
+            id="img"
+            name="img"
           />
 
           <button

frontend/src/services/fileUpload.js

@@ -2,7 +2,7 @@ const baseUrl = import.meta.env.VITE_BACKEND_URL;
 
 const uploadFiles = async (formData) => {
   try {
-    const response = await fetch(baseUrl, {
+    const response = await fetch(`${baseUrl}/v1/uploadfiles`, {
       method: "POST",
       body: formData,
     });