diff --git a/M2_lookuptable_level4_type2.mat b/M2_lookuptable_level4_type2.mat
new file mode 100644
index 0000000..695c692
Binary files /dev/null and b/M2_lookuptable_level4_type2.mat differ
diff --git a/level5_to_level4_M2_type2.m b/level5_to_level4_M2_type2.m
new file mode 100644
index 0000000..03c30f1
--- /dev/null
+++ b/level5_to_level4_M2_type2.m
@@ -0,0 +1,105 @@
+%This programme trace back ROI information based on M2 segmention via a
+%lookuptable.this programme can also generate a new lookuptable.
+%Author: Challen zhangchanglehit@163.com
+%Date: Jun-14th-2017
+clc;
+clear all;
+close all;
+
+load('M2lookuptable.mat'); %para named 'a'
+
+b = a(:,8);
+hashset = unique(b);
+
+res = a;
+row_num = zeros(71,1);
+cnt = 1;
+for z =1:283
+%     flag = 0;
+    dim = size(hashset);
+    if(dim(2)~=0)
+        for j = 1:dim(1)
+            if isequal(res(z,8),hashset(j))
+%                 flag = 1;
+                row_num(cnt)=z;
+                hashset(j)=[];
+                cnt=cnt+1;
+                break;
+            end
+        end
+    end
+%     if(flag==0)
+%         res(i,:)=[];
+%     end
+end
+%after that we got important row num in array row_num
+% 
+cnt = 71;
+for z=283:-1:1
+    if(z~=row_num(cnt))
+        res(z,:)=[];
+    else
+        cnt=cnt-1;
+    end
+end
+%开始进行回溯
+path = 'G:\深圳人民医院\朱进医生\分割结果_M2';
+cd(path)
+dirs_G1=dir([path, '\*.*']);
+dircell_G1=struct2cell(dirs_G1)' ;
+filenames_G1=dircell_G1(3:length(dircell_G1),1);
+for m=1:length(filenames_G1)
+    floder_name = char(filenames_G1(m));
+    cd(floder_name);
+    disp(floder_name);
+    cd('output');
+    dirs_G2=dir;
+    dircell_G2=struct2cell(dirs_G2)' ;
+    filenames_G2=dircell_G2(3:length(dircell_G2),1);
+    len = length(filenames_G2);
+    for j = 1:len
+        
+        filenamesIS = char(filenames_G2(j));
+        jamar = strfind(filenamesIS,'283Labels_M2_MNI.img');
+        if length(jamar) ~= 0
+            
+            [hdr,filetype,fileprefix,machine] = load_nii_hdr(filenamesIS);
+            [img,hdr] = load_nii_img(hdr,filetype,fileprefix,machine);
+            %开始回溯步骤
+            for indexer=1:71
+                for img_index = 1:283
+                    if isequal(res(indexer,8),a(img_index,8))
+                         if indexer~= 43            %43是xxxx对应的区域
+                            index=(img==img_index); %index是图像中对应label number的体素序号
+                            for t=img_index
+                                index1=(img==t);
+                                index=index|index1;
+                            end
+                            img(index)=indexer;
+                         else
+                             index=(img==img_index); %index是图像中对应label number的体素序号
+                            for t=img_index
+                                index1=(img==t);
+                                index=index|index1;
+                            end
+                            img(index)=0;
+                         end
+                    end
+                end
+            end
+            %从1到70排列，因为总共有70个脑区
+            Lables=unique(img);
+            for z=1:71
+                temp=Lables(z); %label number
+                index=(img==temp); %index是图像中对应label number的体素序号
+                img(index)=z-1;
+            end
+            
+            CC_286_L=make_ana(img);     %savemat
+            savename = strcat(floder_name,'_M2_MNI_level4_type2');
+            save_untouch_nii(CC_286_L,savename);
+            cd(path);    
+        end
+    end
+end
+