Robust feature selection to predict tumor treatment outcome
Recurrence of cancer after treatment increases the risk of death. The ability to predict the treatment outcome can help to design the treatment planning and can thus be beneficial to the patient. We aim to select predictive features from clinical and PET (positron emission tomography) based features...
Ausführliche Beschreibung
Gespeichert in:
| Autor*in: |
Mi, Hongmei [verfasserIn] |
|---|
| Format: |
Artikel |
|---|---|
| Sprache: |
Englisch |
| Erschienen: |
2015 |
|---|
| Rechteinformationen: |
Nutzungsrecht: Copyright © 2015 Elsevier B.V. All rights reserved. |
|---|
| Übergeordnetes Werk: |
Enthalten in: Artificial intelligence in medicine - Amsterdam : Elsevier, 1989, 64(2015), 3, Seite 195 |
|---|---|
| Übergeordnetes Werk: |
volume:64 ; year:2015 ; number:3 ; pages:195 |
| Links: |
|---|
| DOI / URN: |
10.1016/j.artmed.2015.07.002 |
|---|
| Katalog-ID: |
OLC195810678X |
|---|
| LEADER | 01000caa a2200265 4500 | ||
|---|---|---|---|
| 001 | OLC195810678X | ||
| 003 | DE-627 | ||
| 005 | 20230508082755.0 | ||
| 007 | tu | ||
| 008 | 160206s2015 xx ||||| 00| ||eng c | ||
| 024 | 7 | |a 10.1016/j.artmed.2015.07.002 |2 doi | |
| 028 | 5 | 2 | |a PQ20160617 |
| 035 | |a (DE-627)OLC195810678X | ||
| 035 | |a (DE-599)GBVOLC195810678X | ||
| 035 | |a (PRQ)c1574-70fa726367eea568164e358ddf2b90d744965b9b40e8ca383822a3c5c15531cf0 | ||
| 035 | |a (KEY)0171479120150000064000300195robustfeatureselectiontopredicttumortreatmentoutco | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 082 | 0 | 4 | |a 33 |a 28 |a 610 |q DNB |
| 082 | 0 | 4 | |a 004 |q AVZ |
| 084 | |a 44.32 |2 bkl | ||
| 084 | |a 54.72 |2 bkl | ||
| 100 | 1 | |a Mi, Hongmei |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Robust feature selection to predict tumor treatment outcome |
| 264 | 1 | |c 2015 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ohne Hilfsmittel zu benutzen |b n |2 rdamedia | ||
| 338 | |a Band |b nc |2 rdacarrier | ||
| 520 | |a Recurrence of cancer after treatment increases the risk of death. The ability to predict the treatment outcome can help to design the treatment planning and can thus be beneficial to the patient. We aim to select predictive features from clinical and PET (positron emission tomography) based features, in order to provide doctors with informative factors so as to anticipate the outcome of the patient treatment. In order to overcome the small sample size problem of datasets usually met in the medical domain, we propose a novel wrapper feature selection algorithm, named HFS (hierarchical forward selection), which searches forward in a hierarchical feature subset space. Feature subsets are iteratively evaluated with the prediction performance using SVM (support vector machine). All feature subsets performing better than those at the preceding iteration are retained. Moreover, as SUV (standardized uptake value) based features have been recognized as significant predictive factors for a patient outcome, we propose to incorporate this prior knowledge into the selection procedure to improve its robustness and reduce its computational cost. Two real-world datasets from cancer patients are included in the evaluation. We extract dozens of clinical and PET-based features to characterize the patient's state, including SUV parameters and texture features. We use leave-one-out cross-validation to evaluate the prediction performance, in terms of prediction accuracy and robustness. Using SVM as the classifier, our HFS method produces accuracy values of 100% and 94% on the two datasets, respectively, and robustness values of 89% and 96%. Without accuracy loss, the prior-based version (pHFS) improves the robustness up to 100% and 98% on the two datasets, respectively. Compared with other feature selection methods, the proposed HFS and pHFS provide the most promising results. For our HFS method, we have empirically shown that the addition of prior knowledge improves the robustness and accelerates the convergence. | ||
| 540 | |a Nutzungsrecht: Copyright © 2015 Elsevier B.V. All rights reserved. | ||
| 700 | 1 | |a Petitjean, Caroline |4 oth | |
| 700 | 1 | |a Dubray, Bernard |4 oth | |
| 700 | 1 | |a Vera, Pierre |4 oth | |
| 700 | 1 | |a Ruan, Su |4 oth | |
| 773 | 0 | 8 | |i Enthalten in |t Artificial intelligence in medicine |d Amsterdam : Elsevier, 1989 |g 64(2015), 3, Seite 195 |w (DE-627)130437190 |w (DE-600)645179-2 |w (DE-576)025083023 |x 0933-3657 |7 nnns |
| 773 | 1 | 8 | |g volume:64 |g year:2015 |g number:3 |g pages:195 |
| 856 | 4 | 1 | |u http://dx.doi.org/10.1016/j.artmed.2015.07.002 |3 Volltext |
| 856 | 4 | 2 | |u http://www.ncbi.nlm.nih.gov/pubmed/26303106 |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a SYSFLAG_A | ||
| 912 | |a GBV_OLC | ||
| 912 | |a SSG-OLC-MAT | ||
| 912 | |a SSG-OLC-PHA | ||
| 912 | |a SSG-OLC-DE-84 | ||
| 912 | |a SSG-OPC-MAT | ||
| 912 | |a GBV_ILN_70 | ||
| 912 | |a GBV_ILN_105 | ||
| 912 | |a GBV_ILN_4112 | ||
| 936 | b | k | |a 44.32 |q AVZ |
| 936 | b | k | |a 54.72 |q AVZ |
| 951 | |a AR | ||
| 952 | |d 64 |j 2015 |e 3 |h 195 | ||
| author_variant |
h m hm |
|---|---|
| matchkey_str |
article:09333657:2015----::outetrslcinordctmrr |
| hierarchy_sort_str |
2015 |
| bklnumber |
44.32 54.72 |
| publishDate |
2015 |
| allfields |
10.1016/j.artmed.2015.07.002 doi PQ20160617 (DE-627)OLC195810678X (DE-599)GBVOLC195810678X (PRQ)c1574-70fa726367eea568164e358ddf2b90d744965b9b40e8ca383822a3c5c15531cf0 (KEY)0171479120150000064000300195robustfeatureselectiontopredicttumortreatmentoutco DE-627 ger DE-627 rakwb eng 33 28 610 DNB 004 AVZ 44.32 bkl 54.72 bkl Mi, Hongmei verfasserin aut Robust feature selection to predict tumor treatment outcome 2015 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier Recurrence of cancer after treatment increases the risk of death. The ability to predict the treatment outcome can help to design the treatment planning and can thus be beneficial to the patient. We aim to select predictive features from clinical and PET (positron emission tomography) based features, in order to provide doctors with informative factors so as to anticipate the outcome of the patient treatment. In order to overcome the small sample size problem of datasets usually met in the medical domain, we propose a novel wrapper feature selection algorithm, named HFS (hierarchical forward selection), which searches forward in a hierarchical feature subset space. Feature subsets are iteratively evaluated with the prediction performance using SVM (support vector machine). All feature subsets performing better than those at the preceding iteration are retained. Moreover, as SUV (standardized uptake value) based features have been recognized as significant predictive factors for a patient outcome, we propose to incorporate this prior knowledge into the selection procedure to improve its robustness and reduce its computational cost. Two real-world datasets from cancer patients are included in the evaluation. We extract dozens of clinical and PET-based features to characterize the patient's state, including SUV parameters and texture features. We use leave-one-out cross-validation to evaluate the prediction performance, in terms of prediction accuracy and robustness. Using SVM as the classifier, our HFS method produces accuracy values of 100% and 94% on the two datasets, respectively, and robustness values of 89% and 96%. Without accuracy loss, the prior-based version (pHFS) improves the robustness up to 100% and 98% on the two datasets, respectively. Compared with other feature selection methods, the proposed HFS and pHFS provide the most promising results. For our HFS method, we have empirically shown that the addition of prior knowledge improves the robustness and accelerates the convergence. Nutzungsrecht: Copyright © 2015 Elsevier B.V. All rights reserved. Petitjean, Caroline oth Dubray, Bernard oth Vera, Pierre oth Ruan, Su oth Enthalten in Artificial intelligence in medicine Amsterdam : Elsevier, 1989 64(2015), 3, Seite 195 (DE-627)130437190 (DE-600)645179-2 (DE-576)025083023 0933-3657 nnns volume:64 year:2015 number:3 pages:195 http://dx.doi.org/10.1016/j.artmed.2015.07.002 Volltext http://www.ncbi.nlm.nih.gov/pubmed/26303106 GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-MAT SSG-OLC-PHA SSG-OLC-DE-84 SSG-OPC-MAT GBV_ILN_70 GBV_ILN_105 GBV_ILN_4112 44.32 AVZ 54.72 AVZ AR 64 2015 3 195 |
| spelling |
10.1016/j.artmed.2015.07.002 doi PQ20160617 (DE-627)OLC195810678X (DE-599)GBVOLC195810678X (PRQ)c1574-70fa726367eea568164e358ddf2b90d744965b9b40e8ca383822a3c5c15531cf0 (KEY)0171479120150000064000300195robustfeatureselectiontopredicttumortreatmentoutco DE-627 ger DE-627 rakwb eng 33 28 610 DNB 004 AVZ 44.32 bkl 54.72 bkl Mi, Hongmei verfasserin aut Robust feature selection to predict tumor treatment outcome 2015 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier Recurrence of cancer after treatment increases the risk of death. The ability to predict the treatment outcome can help to design the treatment planning and can thus be beneficial to the patient. We aim to select predictive features from clinical and PET (positron emission tomography) based features, in order to provide doctors with informative factors so as to anticipate the outcome of the patient treatment. In order to overcome the small sample size problem of datasets usually met in the medical domain, we propose a novel wrapper feature selection algorithm, named HFS (hierarchical forward selection), which searches forward in a hierarchical feature subset space. Feature subsets are iteratively evaluated with the prediction performance using SVM (support vector machine). All feature subsets performing better than those at the preceding iteration are retained. Moreover, as SUV (standardized uptake value) based features have been recognized as significant predictive factors for a patient outcome, we propose to incorporate this prior knowledge into the selection procedure to improve its robustness and reduce its computational cost. Two real-world datasets from cancer patients are included in the evaluation. We extract dozens of clinical and PET-based features to characterize the patient's state, including SUV parameters and texture features. We use leave-one-out cross-validation to evaluate the prediction performance, in terms of prediction accuracy and robustness. Using SVM as the classifier, our HFS method produces accuracy values of 100% and 94% on the two datasets, respectively, and robustness values of 89% and 96%. Without accuracy loss, the prior-based version (pHFS) improves the robustness up to 100% and 98% on the two datasets, respectively. Compared with other feature selection methods, the proposed HFS and pHFS provide the most promising results. For our HFS method, we have empirically shown that the addition of prior knowledge improves the robustness and accelerates the convergence. Nutzungsrecht: Copyright © 2015 Elsevier B.V. All rights reserved. Petitjean, Caroline oth Dubray, Bernard oth Vera, Pierre oth Ruan, Su oth Enthalten in Artificial intelligence in medicine Amsterdam : Elsevier, 1989 64(2015), 3, Seite 195 (DE-627)130437190 (DE-600)645179-2 (DE-576)025083023 0933-3657 nnns volume:64 year:2015 number:3 pages:195 http://dx.doi.org/10.1016/j.artmed.2015.07.002 Volltext http://www.ncbi.nlm.nih.gov/pubmed/26303106 GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-MAT SSG-OLC-PHA SSG-OLC-DE-84 SSG-OPC-MAT GBV_ILN_70 GBV_ILN_105 GBV_ILN_4112 44.32 AVZ 54.72 AVZ AR 64 2015 3 195 |
| allfields_unstemmed |
10.1016/j.artmed.2015.07.002 doi PQ20160617 (DE-627)OLC195810678X (DE-599)GBVOLC195810678X (PRQ)c1574-70fa726367eea568164e358ddf2b90d744965b9b40e8ca383822a3c5c15531cf0 (KEY)0171479120150000064000300195robustfeatureselectiontopredicttumortreatmentoutco DE-627 ger DE-627 rakwb eng 33 28 610 DNB 004 AVZ 44.32 bkl 54.72 bkl Mi, Hongmei verfasserin aut Robust feature selection to predict tumor treatment outcome 2015 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier Recurrence of cancer after treatment increases the risk of death. The ability to predict the treatment outcome can help to design the treatment planning and can thus be beneficial to the patient. We aim to select predictive features from clinical and PET (positron emission tomography) based features, in order to provide doctors with informative factors so as to anticipate the outcome of the patient treatment. In order to overcome the small sample size problem of datasets usually met in the medical domain, we propose a novel wrapper feature selection algorithm, named HFS (hierarchical forward selection), which searches forward in a hierarchical feature subset space. Feature subsets are iteratively evaluated with the prediction performance using SVM (support vector machine). All feature subsets performing better than those at the preceding iteration are retained. Moreover, as SUV (standardized uptake value) based features have been recognized as significant predictive factors for a patient outcome, we propose to incorporate this prior knowledge into the selection procedure to improve its robustness and reduce its computational cost. Two real-world datasets from cancer patients are included in the evaluation. We extract dozens of clinical and PET-based features to characterize the patient's state, including SUV parameters and texture features. We use leave-one-out cross-validation to evaluate the prediction performance, in terms of prediction accuracy and robustness. Using SVM as the classifier, our HFS method produces accuracy values of 100% and 94% on the two datasets, respectively, and robustness values of 89% and 96%. Without accuracy loss, the prior-based version (pHFS) improves the robustness up to 100% and 98% on the two datasets, respectively. Compared with other feature selection methods, the proposed HFS and pHFS provide the most promising results. For our HFS method, we have empirically shown that the addition of prior knowledge improves the robustness and accelerates the convergence. Nutzungsrecht: Copyright © 2015 Elsevier B.V. All rights reserved. Petitjean, Caroline oth Dubray, Bernard oth Vera, Pierre oth Ruan, Su oth Enthalten in Artificial intelligence in medicine Amsterdam : Elsevier, 1989 64(2015), 3, Seite 195 (DE-627)130437190 (DE-600)645179-2 (DE-576)025083023 0933-3657 nnns volume:64 year:2015 number:3 pages:195 http://dx.doi.org/10.1016/j.artmed.2015.07.002 Volltext http://www.ncbi.nlm.nih.gov/pubmed/26303106 GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-MAT SSG-OLC-PHA SSG-OLC-DE-84 SSG-OPC-MAT GBV_ILN_70 GBV_ILN_105 GBV_ILN_4112 44.32 AVZ 54.72 AVZ AR 64 2015 3 195 |
| allfieldsGer |
10.1016/j.artmed.2015.07.002 doi PQ20160617 (DE-627)OLC195810678X (DE-599)GBVOLC195810678X (PRQ)c1574-70fa726367eea568164e358ddf2b90d744965b9b40e8ca383822a3c5c15531cf0 (KEY)0171479120150000064000300195robustfeatureselectiontopredicttumortreatmentoutco DE-627 ger DE-627 rakwb eng 33 28 610 DNB 004 AVZ 44.32 bkl 54.72 bkl Mi, Hongmei verfasserin aut Robust feature selection to predict tumor treatment outcome 2015 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier Recurrence of cancer after treatment increases the risk of death. The ability to predict the treatment outcome can help to design the treatment planning and can thus be beneficial to the patient. We aim to select predictive features from clinical and PET (positron emission tomography) based features, in order to provide doctors with informative factors so as to anticipate the outcome of the patient treatment. In order to overcome the small sample size problem of datasets usually met in the medical domain, we propose a novel wrapper feature selection algorithm, named HFS (hierarchical forward selection), which searches forward in a hierarchical feature subset space. Feature subsets are iteratively evaluated with the prediction performance using SVM (support vector machine). All feature subsets performing better than those at the preceding iteration are retained. Moreover, as SUV (standardized uptake value) based features have been recognized as significant predictive factors for a patient outcome, we propose to incorporate this prior knowledge into the selection procedure to improve its robustness and reduce its computational cost. Two real-world datasets from cancer patients are included in the evaluation. We extract dozens of clinical and PET-based features to characterize the patient's state, including SUV parameters and texture features. We use leave-one-out cross-validation to evaluate the prediction performance, in terms of prediction accuracy and robustness. Using SVM as the classifier, our HFS method produces accuracy values of 100% and 94% on the two datasets, respectively, and robustness values of 89% and 96%. Without accuracy loss, the prior-based version (pHFS) improves the robustness up to 100% and 98% on the two datasets, respectively. Compared with other feature selection methods, the proposed HFS and pHFS provide the most promising results. For our HFS method, we have empirically shown that the addition of prior knowledge improves the robustness and accelerates the convergence. Nutzungsrecht: Copyright © 2015 Elsevier B.V. All rights reserved. Petitjean, Caroline oth Dubray, Bernard oth Vera, Pierre oth Ruan, Su oth Enthalten in Artificial intelligence in medicine Amsterdam : Elsevier, 1989 64(2015), 3, Seite 195 (DE-627)130437190 (DE-600)645179-2 (DE-576)025083023 0933-3657 nnns volume:64 year:2015 number:3 pages:195 http://dx.doi.org/10.1016/j.artmed.2015.07.002 Volltext http://www.ncbi.nlm.nih.gov/pubmed/26303106 GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-MAT SSG-OLC-PHA SSG-OLC-DE-84 SSG-OPC-MAT GBV_ILN_70 GBV_ILN_105 GBV_ILN_4112 44.32 AVZ 54.72 AVZ AR 64 2015 3 195 |
| allfieldsSound |
10.1016/j.artmed.2015.07.002 doi PQ20160617 (DE-627)OLC195810678X (DE-599)GBVOLC195810678X (PRQ)c1574-70fa726367eea568164e358ddf2b90d744965b9b40e8ca383822a3c5c15531cf0 (KEY)0171479120150000064000300195robustfeatureselectiontopredicttumortreatmentoutco DE-627 ger DE-627 rakwb eng 33 28 610 DNB 004 AVZ 44.32 bkl 54.72 bkl Mi, Hongmei verfasserin aut Robust feature selection to predict tumor treatment outcome 2015 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier Recurrence of cancer after treatment increases the risk of death. The ability to predict the treatment outcome can help to design the treatment planning and can thus be beneficial to the patient. We aim to select predictive features from clinical and PET (positron emission tomography) based features, in order to provide doctors with informative factors so as to anticipate the outcome of the patient treatment. In order to overcome the small sample size problem of datasets usually met in the medical domain, we propose a novel wrapper feature selection algorithm, named HFS (hierarchical forward selection), which searches forward in a hierarchical feature subset space. Feature subsets are iteratively evaluated with the prediction performance using SVM (support vector machine). All feature subsets performing better than those at the preceding iteration are retained. Moreover, as SUV (standardized uptake value) based features have been recognized as significant predictive factors for a patient outcome, we propose to incorporate this prior knowledge into the selection procedure to improve its robustness and reduce its computational cost. Two real-world datasets from cancer patients are included in the evaluation. We extract dozens of clinical and PET-based features to characterize the patient's state, including SUV parameters and texture features. We use leave-one-out cross-validation to evaluate the prediction performance, in terms of prediction accuracy and robustness. Using SVM as the classifier, our HFS method produces accuracy values of 100% and 94% on the two datasets, respectively, and robustness values of 89% and 96%. Without accuracy loss, the prior-based version (pHFS) improves the robustness up to 100% and 98% on the two datasets, respectively. Compared with other feature selection methods, the proposed HFS and pHFS provide the most promising results. For our HFS method, we have empirically shown that the addition of prior knowledge improves the robustness and accelerates the convergence. Nutzungsrecht: Copyright © 2015 Elsevier B.V. All rights reserved. Petitjean, Caroline oth Dubray, Bernard oth Vera, Pierre oth Ruan, Su oth Enthalten in Artificial intelligence in medicine Amsterdam : Elsevier, 1989 64(2015), 3, Seite 195 (DE-627)130437190 (DE-600)645179-2 (DE-576)025083023 0933-3657 nnns volume:64 year:2015 number:3 pages:195 http://dx.doi.org/10.1016/j.artmed.2015.07.002 Volltext http://www.ncbi.nlm.nih.gov/pubmed/26303106 GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-MAT SSG-OLC-PHA SSG-OLC-DE-84 SSG-OPC-MAT GBV_ILN_70 GBV_ILN_105 GBV_ILN_4112 44.32 AVZ 54.72 AVZ AR 64 2015 3 195 |
| language |
English |
| source |
Enthalten in Artificial intelligence in medicine 64(2015), 3, Seite 195 volume:64 year:2015 number:3 pages:195 |
| sourceStr |
Enthalten in Artificial intelligence in medicine 64(2015), 3, Seite 195 volume:64 year:2015 number:3 pages:195 |
| format_phy_str_mv |
Article |
| institution |
findex.gbv.de |
| dewey-raw |
33 |
| isfreeaccess_bool |
false |
| container_title |
Artificial intelligence in medicine |
| authorswithroles_txt_mv |
Mi, Hongmei @@aut@@ Petitjean, Caroline @@oth@@ Dubray, Bernard @@oth@@ Vera, Pierre @@oth@@ Ruan, Su @@oth@@ |
| publishDateDaySort_date |
2015-01-01T00:00:00Z |
| hierarchy_top_id |
130437190 |
| dewey-sort |
233 |
| id |
OLC195810678X |
| language_de |
englisch |
| fullrecord |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a2200265 4500</leader><controlfield tag="001">OLC195810678X</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230508082755.0</controlfield><controlfield tag="007">tu</controlfield><controlfield tag="008">160206s2015 xx ||||| 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1016/j.artmed.2015.07.002</subfield><subfield code="2">doi</subfield></datafield><datafield tag="028" ind1="5" ind2="2"><subfield code="a">PQ20160617</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)OLC195810678X</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)GBVOLC195810678X</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(PRQ)c1574-70fa726367eea568164e358ddf2b90d744965b9b40e8ca383822a3c5c15531cf0</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(KEY)0171479120150000064000300195robustfeatureselectiontopredicttumortreatmentoutco</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">33</subfield><subfield code="a">28</subfield><subfield code="a">610</subfield><subfield code="q">DNB</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">004</subfield><subfield code="q">AVZ</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">44.32</subfield><subfield code="2">bkl</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">54.72</subfield><subfield code="2">bkl</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Mi, Hongmei</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Robust feature selection to predict tumor treatment outcome</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2015</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">ohne Hilfsmittel zu benutzen</subfield><subfield code="b">n</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Band</subfield><subfield code="b">nc</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Recurrence of cancer after treatment increases the risk of death. The ability to predict the treatment outcome can help to design the treatment planning and can thus be beneficial to the patient. We aim to select predictive features from clinical and PET (positron emission tomography) based features, in order to provide doctors with informative factors so as to anticipate the outcome of the patient treatment. In order to overcome the small sample size problem of datasets usually met in the medical domain, we propose a novel wrapper feature selection algorithm, named HFS (hierarchical forward selection), which searches forward in a hierarchical feature subset space. Feature subsets are iteratively evaluated with the prediction performance using SVM (support vector machine). All feature subsets performing better than those at the preceding iteration are retained. Moreover, as SUV (standardized uptake value) based features have been recognized as significant predictive factors for a patient outcome, we propose to incorporate this prior knowledge into the selection procedure to improve its robustness and reduce its computational cost. Two real-world datasets from cancer patients are included in the evaluation. We extract dozens of clinical and PET-based features to characterize the patient's state, including SUV parameters and texture features. We use leave-one-out cross-validation to evaluate the prediction performance, in terms of prediction accuracy and robustness. Using SVM as the classifier, our HFS method produces accuracy values of 100% and 94% on the two datasets, respectively, and robustness values of 89% and 96%. Without accuracy loss, the prior-based version (pHFS) improves the robustness up to 100% and 98% on the two datasets, respectively. Compared with other feature selection methods, the proposed HFS and pHFS provide the most promising results. For our HFS method, we have empirically shown that the addition of prior knowledge improves the robustness and accelerates the convergence.</subfield></datafield><datafield tag="540" ind1=" " ind2=" "><subfield code="a">Nutzungsrecht: Copyright © 2015 Elsevier B.V. All rights reserved.</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Petitjean, Caroline</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Dubray, Bernard</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Vera, Pierre</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Ruan, Su</subfield><subfield code="4">oth</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="t">Artificial intelligence in medicine</subfield><subfield code="d">Amsterdam : Elsevier, 1989</subfield><subfield code="g">64(2015), 3, Seite 195</subfield><subfield code="w">(DE-627)130437190</subfield><subfield code="w">(DE-600)645179-2</subfield><subfield code="w">(DE-576)025083023</subfield><subfield code="x">0933-3657</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:64</subfield><subfield code="g">year:2015</subfield><subfield code="g">number:3</subfield><subfield code="g">pages:195</subfield></datafield><datafield tag="856" ind1="4" ind2="1"><subfield code="u">http://dx.doi.org/10.1016/j.artmed.2015.07.002</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="856" ind1="4" ind2="2"><subfield code="u">http://www.ncbi.nlm.nih.gov/pubmed/26303106</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_OLC</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-MAT</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-PHA</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-DE-84</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OPC-MAT</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_70</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_105</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4112</subfield></datafield><datafield tag="936" ind1="b" ind2="k"><subfield code="a">44.32</subfield><subfield code="q">AVZ</subfield></datafield><datafield tag="936" ind1="b" ind2="k"><subfield code="a">54.72</subfield><subfield code="q">AVZ</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">64</subfield><subfield code="j">2015</subfield><subfield code="e">3</subfield><subfield code="h">195</subfield></datafield></record></collection>
|
| author |
Mi, Hongmei |
| spellingShingle |
Mi, Hongmei ddc 33 ddc 004 bkl 44.32 bkl 54.72 Robust feature selection to predict tumor treatment outcome |
| authorStr |
Mi, Hongmei |
| ppnlink_with_tag_str_mv |
@@773@@(DE-627)130437190 |
| format |
Article |
| dewey-ones |
033 - Encyclopedias in other Germanic languages 028 - Reading & use of other information media 610 - Medicine & health 004 - Data processing & computer science |
| delete_txt_mv |
keep |
| author_role |
aut |
| collection |
OLC |
| remote_str |
false |
| illustrated |
Not Illustrated |
| issn |
0933-3657 |
| topic_title |
33 28 610 DNB 004 AVZ 44.32 bkl 54.72 bkl Robust feature selection to predict tumor treatment outcome |
| topic |
ddc 33 ddc 004 bkl 44.32 bkl 54.72 |
| topic_unstemmed |
ddc 33 ddc 004 bkl 44.32 bkl 54.72 |
| topic_browse |
ddc 33 ddc 004 bkl 44.32 bkl 54.72 |
| format_facet |
Aufsätze Gedruckte Aufsätze |
| format_main_str_mv |
Text Zeitschrift/Artikel |
| carriertype_str_mv |
nc |
| author2_variant |
c p cp b d bd p v pv s r sr |
| hierarchy_parent_title |
Artificial intelligence in medicine |
| hierarchy_parent_id |
130437190 |
| dewey-tens |
030 - Encyclopedias & books of facts 020 - Library & information sciences 610 - Medicine & health 000 - Computer science, knowledge & systems |
| hierarchy_top_title |
Artificial intelligence in medicine |
| isfreeaccess_txt |
false |
| familylinks_str_mv |
(DE-627)130437190 (DE-600)645179-2 (DE-576)025083023 |
| title |
Robust feature selection to predict tumor treatment outcome |
| ctrlnum |
(DE-627)OLC195810678X (DE-599)GBVOLC195810678X (PRQ)c1574-70fa726367eea568164e358ddf2b90d744965b9b40e8ca383822a3c5c15531cf0 (KEY)0171479120150000064000300195robustfeatureselectiontopredicttumortreatmentoutco |
| title_full |
Robust feature selection to predict tumor treatment outcome |
| author_sort |
Mi, Hongmei |
| journal |
Artificial intelligence in medicine |
| journalStr |
Artificial intelligence in medicine |
| lang_code |
eng |
| isOA_bool |
false |
| dewey-hundreds |
000 - Computer science, information & general works 600 - Technology |
| recordtype |
marc |
| publishDateSort |
2015 |
| contenttype_str_mv |
txt |
| container_start_page |
195 |
| author_browse |
Mi, Hongmei |
| container_volume |
64 |
| class |
33 28 610 DNB 004 AVZ 44.32 bkl 54.72 bkl |
| format_se |
Aufsätze |
| author-letter |
Mi, Hongmei |
| doi_str_mv |
10.1016/j.artmed.2015.07.002 |
| dewey-full |
33 28 610 004 |
| title_sort |
robust feature selection to predict tumor treatment outcome |
| title_auth |
Robust feature selection to predict tumor treatment outcome |
| abstract |
Recurrence of cancer after treatment increases the risk of death. The ability to predict the treatment outcome can help to design the treatment planning and can thus be beneficial to the patient. We aim to select predictive features from clinical and PET (positron emission tomography) based features, in order to provide doctors with informative factors so as to anticipate the outcome of the patient treatment. In order to overcome the small sample size problem of datasets usually met in the medical domain, we propose a novel wrapper feature selection algorithm, named HFS (hierarchical forward selection), which searches forward in a hierarchical feature subset space. Feature subsets are iteratively evaluated with the prediction performance using SVM (support vector machine). All feature subsets performing better than those at the preceding iteration are retained. Moreover, as SUV (standardized uptake value) based features have been recognized as significant predictive factors for a patient outcome, we propose to incorporate this prior knowledge into the selection procedure to improve its robustness and reduce its computational cost. Two real-world datasets from cancer patients are included in the evaluation. We extract dozens of clinical and PET-based features to characterize the patient's state, including SUV parameters and texture features. We use leave-one-out cross-validation to evaluate the prediction performance, in terms of prediction accuracy and robustness. Using SVM as the classifier, our HFS method produces accuracy values of 100% and 94% on the two datasets, respectively, and robustness values of 89% and 96%. Without accuracy loss, the prior-based version (pHFS) improves the robustness up to 100% and 98% on the two datasets, respectively. Compared with other feature selection methods, the proposed HFS and pHFS provide the most promising results. For our HFS method, we have empirically shown that the addition of prior knowledge improves the robustness and accelerates the convergence. |
| abstractGer |
Recurrence of cancer after treatment increases the risk of death. The ability to predict the treatment outcome can help to design the treatment planning and can thus be beneficial to the patient. We aim to select predictive features from clinical and PET (positron emission tomography) based features, in order to provide doctors with informative factors so as to anticipate the outcome of the patient treatment. In order to overcome the small sample size problem of datasets usually met in the medical domain, we propose a novel wrapper feature selection algorithm, named HFS (hierarchical forward selection), which searches forward in a hierarchical feature subset space. Feature subsets are iteratively evaluated with the prediction performance using SVM (support vector machine). All feature subsets performing better than those at the preceding iteration are retained. Moreover, as SUV (standardized uptake value) based features have been recognized as significant predictive factors for a patient outcome, we propose to incorporate this prior knowledge into the selection procedure to improve its robustness and reduce its computational cost. Two real-world datasets from cancer patients are included in the evaluation. We extract dozens of clinical and PET-based features to characterize the patient's state, including SUV parameters and texture features. We use leave-one-out cross-validation to evaluate the prediction performance, in terms of prediction accuracy and robustness. Using SVM as the classifier, our HFS method produces accuracy values of 100% and 94% on the two datasets, respectively, and robustness values of 89% and 96%. Without accuracy loss, the prior-based version (pHFS) improves the robustness up to 100% and 98% on the two datasets, respectively. Compared with other feature selection methods, the proposed HFS and pHFS provide the most promising results. For our HFS method, we have empirically shown that the addition of prior knowledge improves the robustness and accelerates the convergence. |
| abstract_unstemmed |
Recurrence of cancer after treatment increases the risk of death. The ability to predict the treatment outcome can help to design the treatment planning and can thus be beneficial to the patient. We aim to select predictive features from clinical and PET (positron emission tomography) based features, in order to provide doctors with informative factors so as to anticipate the outcome of the patient treatment. In order to overcome the small sample size problem of datasets usually met in the medical domain, we propose a novel wrapper feature selection algorithm, named HFS (hierarchical forward selection), which searches forward in a hierarchical feature subset space. Feature subsets are iteratively evaluated with the prediction performance using SVM (support vector machine). All feature subsets performing better than those at the preceding iteration are retained. Moreover, as SUV (standardized uptake value) based features have been recognized as significant predictive factors for a patient outcome, we propose to incorporate this prior knowledge into the selection procedure to improve its robustness and reduce its computational cost. Two real-world datasets from cancer patients are included in the evaluation. We extract dozens of clinical and PET-based features to characterize the patient's state, including SUV parameters and texture features. We use leave-one-out cross-validation to evaluate the prediction performance, in terms of prediction accuracy and robustness. Using SVM as the classifier, our HFS method produces accuracy values of 100% and 94% on the two datasets, respectively, and robustness values of 89% and 96%. Without accuracy loss, the prior-based version (pHFS) improves the robustness up to 100% and 98% on the two datasets, respectively. Compared with other feature selection methods, the proposed HFS and pHFS provide the most promising results. For our HFS method, we have empirically shown that the addition of prior knowledge improves the robustness and accelerates the convergence. |
| collection_details |
GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-MAT SSG-OLC-PHA SSG-OLC-DE-84 SSG-OPC-MAT GBV_ILN_70 GBV_ILN_105 GBV_ILN_4112 |
| container_issue |
3 |
| title_short |
Robust feature selection to predict tumor treatment outcome |
| url |
http://dx.doi.org/10.1016/j.artmed.2015.07.002 http://www.ncbi.nlm.nih.gov/pubmed/26303106 |
| remote_bool |
false |
| author2 |
Petitjean, Caroline Dubray, Bernard Vera, Pierre Ruan, Su |
| author2Str |
Petitjean, Caroline Dubray, Bernard Vera, Pierre Ruan, Su |
| ppnlink |
130437190 |
| mediatype_str_mv |
n |
| isOA_txt |
false |
| hochschulschrift_bool |
false |
| author2_role |
oth oth oth oth |
| doi_str |
10.1016/j.artmed.2015.07.002 |
| up_date |
2024-07-04T02:09:53.367Z |
| _version_ |
1803612579030892544 |
| fullrecord_marcxml |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a2200265 4500</leader><controlfield tag="001">OLC195810678X</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230508082755.0</controlfield><controlfield tag="007">tu</controlfield><controlfield tag="008">160206s2015 xx ||||| 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1016/j.artmed.2015.07.002</subfield><subfield code="2">doi</subfield></datafield><datafield tag="028" ind1="5" ind2="2"><subfield code="a">PQ20160617</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)OLC195810678X</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)GBVOLC195810678X</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(PRQ)c1574-70fa726367eea568164e358ddf2b90d744965b9b40e8ca383822a3c5c15531cf0</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(KEY)0171479120150000064000300195robustfeatureselectiontopredicttumortreatmentoutco</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">33</subfield><subfield code="a">28</subfield><subfield code="a">610</subfield><subfield code="q">DNB</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">004</subfield><subfield code="q">AVZ</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">44.32</subfield><subfield code="2">bkl</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">54.72</subfield><subfield code="2">bkl</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Mi, Hongmei</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Robust feature selection to predict tumor treatment outcome</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2015</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">ohne Hilfsmittel zu benutzen</subfield><subfield code="b">n</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Band</subfield><subfield code="b">nc</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Recurrence of cancer after treatment increases the risk of death. The ability to predict the treatment outcome can help to design the treatment planning and can thus be beneficial to the patient. We aim to select predictive features from clinical and PET (positron emission tomography) based features, in order to provide doctors with informative factors so as to anticipate the outcome of the patient treatment. In order to overcome the small sample size problem of datasets usually met in the medical domain, we propose a novel wrapper feature selection algorithm, named HFS (hierarchical forward selection), which searches forward in a hierarchical feature subset space. Feature subsets are iteratively evaluated with the prediction performance using SVM (support vector machine). All feature subsets performing better than those at the preceding iteration are retained. Moreover, as SUV (standardized uptake value) based features have been recognized as significant predictive factors for a patient outcome, we propose to incorporate this prior knowledge into the selection procedure to improve its robustness and reduce its computational cost. Two real-world datasets from cancer patients are included in the evaluation. We extract dozens of clinical and PET-based features to characterize the patient's state, including SUV parameters and texture features. We use leave-one-out cross-validation to evaluate the prediction performance, in terms of prediction accuracy and robustness. Using SVM as the classifier, our HFS method produces accuracy values of 100% and 94% on the two datasets, respectively, and robustness values of 89% and 96%. Without accuracy loss, the prior-based version (pHFS) improves the robustness up to 100% and 98% on the two datasets, respectively. Compared with other feature selection methods, the proposed HFS and pHFS provide the most promising results. For our HFS method, we have empirically shown that the addition of prior knowledge improves the robustness and accelerates the convergence.</subfield></datafield><datafield tag="540" ind1=" " ind2=" "><subfield code="a">Nutzungsrecht: Copyright © 2015 Elsevier B.V. All rights reserved.</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Petitjean, Caroline</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Dubray, Bernard</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Vera, Pierre</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Ruan, Su</subfield><subfield code="4">oth</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="t">Artificial intelligence in medicine</subfield><subfield code="d">Amsterdam : Elsevier, 1989</subfield><subfield code="g">64(2015), 3, Seite 195</subfield><subfield code="w">(DE-627)130437190</subfield><subfield code="w">(DE-600)645179-2</subfield><subfield code="w">(DE-576)025083023</subfield><subfield code="x">0933-3657</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:64</subfield><subfield code="g">year:2015</subfield><subfield code="g">number:3</subfield><subfield code="g">pages:195</subfield></datafield><datafield tag="856" ind1="4" ind2="1"><subfield code="u">http://dx.doi.org/10.1016/j.artmed.2015.07.002</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="856" ind1="4" ind2="2"><subfield code="u">http://www.ncbi.nlm.nih.gov/pubmed/26303106</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_OLC</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-MAT</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-PHA</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-DE-84</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OPC-MAT</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_70</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_105</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4112</subfield></datafield><datafield tag="936" ind1="b" ind2="k"><subfield code="a">44.32</subfield><subfield code="q">AVZ</subfield></datafield><datafield tag="936" ind1="b" ind2="k"><subfield code="a">54.72</subfield><subfield code="q">AVZ</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">64</subfield><subfield code="j">2015</subfield><subfield code="e">3</subfield><subfield code="h">195</subfield></datafield></record></collection>
|
| score |
7.4015102 |