在matlab中选择重叠数据？

Question

我有一个看起来像这样的表(有更多的行)：

Patient startday    stopday drug
P1      2           4       D-A
P1      3           7       D-B
P1      9           13      D-C
P2      0           6       D-A
P2      2           10      D-C
P2      3           7       D-D
P2      8           12      D-B

在我的matlab代码中，我需要检查在相同的开始/停止天数间隔内是否存在两种相同的药物。例如，我如何陈述患者P2中D-A和D-D的共同假设？换句话说:患者P2在同一天内(0-6和3-7，重叠在3-6上)同时服用了这两种药物，D-A和D-D。我的输出需要是一列1/0，说明这两种药物是否发生了重叠。(代码的最终目标，如果有用的话，是生存分析，这将是一个协变量)

我是一个初学者，我正在考虑的策略是:从一个独特的患者列表中，选择列表中的每个患者，然后选择药物1和药物2，看看开始/停止的天数是否重叠。对这些行写1，对所有其他行写0。

我在寻找各种帮助！

Answer 1

这里有一种方法，我认为它能满足您的需要。

% data setup
Patient = categorical(["P1";"P1";"P1";"P2";"P2";"P2";"P2"]);
startday = [2;3;9;0;2;3;8];
stopday = [4;7;13;6;10;7;12];
drug = categorical(["D-A";"D-B";"D-C";"D-A";"D-C";"D-D";"D-B"]);
t = table(Patient, startday, stopday, drug);

% processing
unique_patients = unique(t.Patient);
unique_drugs = unique(t.drug);
% get each unique pair of drugs (ie. D-A & D-B, D-B & D-C ... D-C & D-D)
drug_pairs = nchoosek(unique_drugs, 2);

% results matrix to store the data in for X unique patients and Y pairs of drugs
results = zeros(length(unique_patients), length(drug_pairs));

for i = 1:length(unique_patients)
    patient = unique_patients(i);
    patient_data = t(t.Patient == patient,:);
    patient_drugs = patient_data.drug;

    for j = 1:length(drug_pairs)
        drug_pair = drug_pairs(j,:);
        % find the data for this patient with each of these drugs
        drug1_data = patient_data(patient_drugs == drug_pair(1),:);
        drug2_data = patient_data(patient_drugs == drug_pair(2),:);
        % size(drugN_data, 1) will be 0 if this patient didn't have this drug
        % so we check to make sure both drugs were administered
        if size(drug1_data, 1) > 0 && size(drug2_data,1) > 0
            % if drug 2 was stopped before drug 1 started or vice versa, there's no overlap
            % so if this is not the case, there is an overlap and we store a 1 in results 
            if ~(drug1_data.stopday < drug2_data.startday || drug2_data.stopday < drug1_data.startday)
                results(i,j) = 1;
            end
        end
    end
end

% generate the names for the columns for each drug pair - can be edited as desired
drug_pair_column_names = [string(drug_pairs(:,1)) + " & " + string(drug_pairs(:,2))];

% construct the final table by joining:
% - first column, the patient names
% - rest of the columns, the results data we calculated above
results_table = [table(unique_patients,'VariableNames',["Patient"]), array2table(results, 'VariableNames', drug_pair_column_names)];

结果：

results_table =

    Patient    D-A & D-B    D-A & D-C    D-A & D-D    D-B & D-C    D-B & D-D    D-C & D-D
    _______    _________    _________    _________    _________    _________    _________

      P1           1            0            0            0            0            0    
      P2           0            1            1            1            0            1