Transfer learning approach based on computed tomography images for predicting late xerostomia after radiotherapy in patients with oropharyngeal cancer
Background and purposeAlthough the latest breakthroughs in radiotherapy (RT) techniques have led to a decrease in adverse event rates, these techniques are still associated with substantial toxicity, including xerostomia. Imaging biomarkers could be useful to predict the toxicity risk related to eac...
| Main Authors: | , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-09-01
|
| Series: | Frontiers in Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmed.2022.993395/full |
| _version_ | 1811319028218593280 |
|---|---|
| author | Annarita Fanizzi Giovanni Scognamillo Alessandra Nestola Santa Bambace Samantha Bove Maria Colomba Comes Cristian Cristofaro Vittorio Didonna Alessia Di Rito Angelo Errico Loredana Palermo Pasquale Tamborra Michele Troiano Salvatore Parisi Rossella Villani Alfredo Zito Marco Lioce Raffaella Massafra |
| author_facet | Annarita Fanizzi Giovanni Scognamillo Alessandra Nestola Santa Bambace Samantha Bove Maria Colomba Comes Cristian Cristofaro Vittorio Didonna Alessia Di Rito Angelo Errico Loredana Palermo Pasquale Tamborra Michele Troiano Salvatore Parisi Rossella Villani Alfredo Zito Marco Lioce Raffaella Massafra |
| author_sort | Annarita Fanizzi |
| collection | DOAJ |
| description | Background and purposeAlthough the latest breakthroughs in radiotherapy (RT) techniques have led to a decrease in adverse event rates, these techniques are still associated with substantial toxicity, including xerostomia. Imaging biomarkers could be useful to predict the toxicity risk related to each individual patient. Our preliminary work aims to develop a radiomic-based support tool exploiting pre-treatment CT images to predict late xerostomia risk in 3 months after RT in patients with oropharyngeal cancer (OPC).Materials and methodsWe performed a multicenter data collection. We enrolled 61 patients referred to three care centers in Apulia, Italy, out of which 22 patients experienced at least mild xerostomia 3 months after the end of the RT cycle. Pre-treatment CT images, clinical and dose features, and alcohol-smoking habits were collected. We proposed a transfer learning approach to extract quantitative imaging features from CT images by means of a pre-trained convolutional neural network (CNN) architecture. An optimal feature subset was then identified to train an SVM classifier. To evaluate the robustness of the proposed model with respect to different manual contouring practices on CTs, we repeated the same image analysis pipeline on “fake” parotid contours.ResultsThe best performances were achieved by the model exploiting the radiomic features alone. On the independent test, the model reached median AUC, accuracy, sensitivity, and specificity values of 81.17, 83.33, 71.43, and 90.91%, respectively. The model was robust with respect to diverse manual parotid contouring procedures.ConclusionRadiomic analysis could help to develop a valid support tool for clinicians in planning radiotherapy treatment, by providing a risk score of the toxicity development for each individual patient, thus improving the quality of life of the same patient, without compromising patient care. |
| first_indexed | 2024-04-13T12:35:10Z |
| format | Article |
| id | doaj.art-315f385a0a2c409c85448f891a5abfd4 |
| institution | Directory Open Access Journal |
| issn | 2296-858X |
| language | English |
| last_indexed | 2024-04-13T12:35:10Z |
| publishDate | 2022-09-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Medicine |
| spelling | doaj.art-315f385a0a2c409c85448f891a5abfd42022-12-22T02:46:40ZengFrontiers Media S.A.Frontiers in Medicine2296-858X2022-09-01910.3389/fmed.2022.993395993395Transfer learning approach based on computed tomography images for predicting late xerostomia after radiotherapy in patients with oropharyngeal cancerAnnarita Fanizzi0Giovanni Scognamillo1Alessandra Nestola2Santa Bambace3Samantha Bove4Maria Colomba Comes5Cristian Cristofaro6Vittorio Didonna7Alessia Di Rito8Angelo Errico9Loredana Palermo10Pasquale Tamborra11Michele Troiano12Salvatore Parisi13Rossella Villani14Alfredo Zito15Marco Lioce16Raffaella Massafra17IRCCS Istituto Tumori “Giovanni Paolo II,”Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II,”Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II,”Bari, ItalyOspedale Monsignor Raffaele Dimiccoli, Barletta, ItalyIRCCS Istituto Tumori “Giovanni Paolo II,”Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II,”Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II,”Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II,”Bari, ItalyOspedale Monsignor Raffaele Dimiccoli, Barletta, ItalyOspedale Monsignor Raffaele Dimiccoli, Barletta, ItalyIRCCS Istituto Tumori “Giovanni Paolo II,”Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II,”Bari, ItalyIRCCS Casa Sollievo della Sofferenza, Opera di San Pio da Pietrelcina Viale Cappuccini, Foggia, ItalyIRCCS Casa Sollievo della Sofferenza, Opera di San Pio da Pietrelcina Viale Cappuccini, Foggia, ItalyIRCCS Istituto Tumori “Giovanni Paolo II,”Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II,”Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II,”Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II,”Bari, ItalyBackground and purposeAlthough the latest breakthroughs in radiotherapy (RT) techniques have led to a decrease in adverse event rates, these techniques are still associated with substantial toxicity, including xerostomia. Imaging biomarkers could be useful to predict the toxicity risk related to each individual patient. Our preliminary work aims to develop a radiomic-based support tool exploiting pre-treatment CT images to predict late xerostomia risk in 3 months after RT in patients with oropharyngeal cancer (OPC).Materials and methodsWe performed a multicenter data collection. We enrolled 61 patients referred to three care centers in Apulia, Italy, out of which 22 patients experienced at least mild xerostomia 3 months after the end of the RT cycle. Pre-treatment CT images, clinical and dose features, and alcohol-smoking habits were collected. We proposed a transfer learning approach to extract quantitative imaging features from CT images by means of a pre-trained convolutional neural network (CNN) architecture. An optimal feature subset was then identified to train an SVM classifier. To evaluate the robustness of the proposed model with respect to different manual contouring practices on CTs, we repeated the same image analysis pipeline on “fake” parotid contours.ResultsThe best performances were achieved by the model exploiting the radiomic features alone. On the independent test, the model reached median AUC, accuracy, sensitivity, and specificity values of 81.17, 83.33, 71.43, and 90.91%, respectively. The model was robust with respect to diverse manual parotid contouring procedures.ConclusionRadiomic analysis could help to develop a valid support tool for clinicians in planning radiotherapy treatment, by providing a risk score of the toxicity development for each individual patient, thus improving the quality of life of the same patient, without compromising patient care.https://www.frontiersin.org/articles/10.3389/fmed.2022.993395/fulldeep learningxerostomiaoropharyngeal cancerCT imagesCNN–convolutional neural network |
| spellingShingle | Annarita Fanizzi Giovanni Scognamillo Alessandra Nestola Santa Bambace Samantha Bove Maria Colomba Comes Cristian Cristofaro Vittorio Didonna Alessia Di Rito Angelo Errico Loredana Palermo Pasquale Tamborra Michele Troiano Salvatore Parisi Rossella Villani Alfredo Zito Marco Lioce Raffaella Massafra Transfer learning approach based on computed tomography images for predicting late xerostomia after radiotherapy in patients with oropharyngeal cancer Frontiers in Medicine deep learning xerostomia oropharyngeal cancer CT images CNN–convolutional neural network |
| title | Transfer learning approach based on computed tomography images for predicting late xerostomia after radiotherapy in patients with oropharyngeal cancer |
| title_full | Transfer learning approach based on computed tomography images for predicting late xerostomia after radiotherapy in patients with oropharyngeal cancer |
| title_fullStr | Transfer learning approach based on computed tomography images for predicting late xerostomia after radiotherapy in patients with oropharyngeal cancer |
| title_full_unstemmed | Transfer learning approach based on computed tomography images for predicting late xerostomia after radiotherapy in patients with oropharyngeal cancer |
| title_short | Transfer learning approach based on computed tomography images for predicting late xerostomia after radiotherapy in patients with oropharyngeal cancer |
| title_sort | transfer learning approach based on computed tomography images for predicting late xerostomia after radiotherapy in patients with oropharyngeal cancer |
| topic | deep learning xerostomia oropharyngeal cancer CT images CNN–convolutional neural network |
| url | https://www.frontiersin.org/articles/10.3389/fmed.2022.993395/full |
| work_keys_str_mv | AT annaritafanizzi transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT giovanniscognamillo transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT alessandranestola transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT santabambace transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT samanthabove transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT mariacolombacomes transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT cristiancristofaro transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT vittoriodidonna transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT alessiadirito transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT angeloerrico transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT loredanapalermo transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT pasqualetamborra transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT micheletroiano transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT salvatoreparisi transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT rossellavillani transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT alfredozito transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT marcolioce transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer AT raffaellamassafra transferlearningapproachbasedoncomputedtomographyimagesforpredictinglatexerostomiaafterradiotherapyinpatientswithoropharyngealcancer |