minors

index.html

@@ -234,18 +234,16 @@ <h2 class="title is-3">Proposed method</h2>
           from pronounced streaking artifacts.
 
         <ol type="a">
-          <li>Initial deep learning techniques applied to post-processing tasks on FBP reconstructed images show
+          <li>Initial deep learning techniques applied FBP reconstructed images as a post-processing task show
             promise in artifact removal and structure preservation but face limitations due to constrained receptive
             fields, leading to suboptimal results. However, these methods are computationally efficient.
           </li>
-          <li>Deep unrolling algorithms, such as the learned primal-dual (LPD) algorithm, have been introduced to
-            optimize the reconstruction process. However, they face issues such as slow convergence and high
-            computational costs. Consequently, there is a need to explore more efficient
-            alternatives due to the difficulties in capturing long-range dependencies and the growing computational
-            demands of modern neural networks.
+          <li>Deep unrolling algorithms, such as the RegFormer algorithm, have been introduced as an iterative reconstruction methods. 
+            However, they face issues such as slow convergence and high computational costs. Consequently, there is a need to explore more efficient
+            alternatives due to the difficulties in capturing long-range dependencies and the growing computational demands of modern neural networks.
           </li>
           <li>The paper introduces a second-order unrolling network called QN-Mixer, which employs a latent
-            BFGS algorithm to approximate the Hessian matrix with a deep-net learned regularization term.
+            BFGS algorithm to approximate the inverse Hessian matrix with a deep-net learned regularization term.
             It outperforms state-of-the-art methods in terms of quantitative metrics while requiring fewer
             iterations than first-order unrolling networks.
           </li>
@@ -269,7 +267,7 @@ <h2 class="title is-3">Methodology</h2>
         <img src="static/images/overview.png" alt="overview" />
 
         <p class="content has-text-justified">
-          Our method is a new type of second-order unrolling network, drawing inspiration from the quasi-Newton method.
+          Our method is a new type of unrolling networks, drawing inspiration from the quasi-Newton method.
           The figure above illustrates the QN-Mixer architecture. It approximates the inverse Hessian matrix using a
           latent BFGS algorithm and incorporates a non-local regularization term, Incept-Mixer, aimed at capturing
           non-local relationships. To address the computational challenges associated with full inverse Hessian matrix
@@ -283,7 +281,7 @@ <h2 class="title is-3">Methodology</h2>
         <p class="content has-text-justified">
           Incept-Mixer bloc is crafted by drawing inspiration from both the multi-layer perceptron mixer and the
           inception architecture, leveraging the strengths of each: capturing long-range interactions through the
-          attention-like mechanism of MLPMixer and extracting local invariant features from the inception block.
+          attention-like mechanism of MLP-Mixer and extracting local invariant features from the inception block.
         </p>