How to use the @tensorflow/tfjs-node.node function in @tensorflow/tfjs-node

* Tensorflow.js Examples for Node.js
 * Script adatapted from 
 * https://github.com/tensorflow/tfjs-examples
 * https://groups.google.com/a/tensorflow.org/forum/#!forum/tfjs
 * @author Loreto Parisi (loretoparisi@gmail.com)
 * @copyright 2020 Loreto Parisi (loretoparisi@gmail.com)
 */

const tf = require('@tensorflow/tfjs-node');

var model, image;
const model_path = './model/new_object_detection_1';
const labels = require('./model/new_object_detection_1/assets/labels.json');
const { createCanvas, Image } = require('canvas');

tf.node.loadSavedModel(model_path, ['serve'], 'serving_default')
    .then(res => {
        model = res; // LP: loaded TFSavedModel
        console.log(model);
        return tf.node.getMetaGraphsFromSavedModel(model_path);
    })
    .then(modelInfo => {
        console.log(modelInfo);
        image = require('fs').readFileSync('./image.jpeg');
        const uint8array = new Uint8Array(image);
        // Decode the image into a tensor.
        return tf.node.decodeImage(uint8array);
    })
    .then(imageTensor => {
        const input = imageTensor.expandDims(0);
        // Feed the image tensor into the model for inference.
        const startTime = tf.util.now();

.then(modelInfo => {
        console.log(modelInfo);
        image = require('fs').readFileSync('./image.jpeg');
        const uint8array = new Uint8Array(image);
        // Decode the image into a tensor.
        return tf.node.decodeImage(uint8array);
    })
    .then(imageTensor => {

async function loadModel() {
  // Warm up the model
  if (!objectDetectionModel) {
    objectDetectionModel = await tf.node.loadSavedModel(
        './model/new_object_detection_1', ['serve'], 'serving_default');
  }
  const tempTensor = tf.zeros([1, 2, 2, 3]).toInt();
  objectDetectionModel.predict(tempTensor);
}

getTensor3dObject(numOfChannels) {

        const imageData = this.inputImage.replace('data:image/jpeg;base64','')
                            .replace('data:image/png;base64','');
        
        const imageArray = toUint8Array(imageData);
        
        const tensor3d = tf.node.decodeJpeg( imageArray, numOfChannels );

        return tensor3d;
    }

busboy.on('finish', async () => {
    const buf = req.files.file[0].buffer;
    const uint8array = new Uint8Array(buf);

    if (!objectDetectionModel) {
      objectDetectionModel = await tf.node.loadSavedModel(
          './model/new_object_detection_1', ['serve'], 'serving_default');
    }
    const imageTensor = await tf.node.decodeImage(uint8array);
    const input = imageTensor.expandDims(0);

    let outputTensor = objectDetectionModel.predict({'x': input});

    const scores = await outputTensor['detection_scores'].arraySync();
    const boxes = await outputTensor['detection_boxes'].arraySync();
    const names = await outputTensor['detection_classes'].arraySync();
    outputTensor['detection_scores'].dispose();
    outputTensor['detection_boxes'].dispose();
    outputTensor['detection_classes'].dispose();
    outputTensor['num_detections'].dispose();
    const detectedBoxes = [];
    const detectedNames = [];
    for (let i = 0; i < scores[0].length; i++) {
      if (scores[0][i] > 0.3) {
        detectedBoxes.push(boxes[0][i]);