Abstract
Purpose
To describe in detail a dataset consisting of longitudinal fan‐beam computed tomography (CT) imaging to visualize anatomical changes in head‐and‐neck squamous cell carcinoma (HNSCC) patients throughout radiotherapy (RT) treatment course.
Acquisition and validation methods
This dataset consists of CT images from 31 HNSCC patients who underwent volumetric modulated arc therapy (VMAT). Patients had three CT scans acquired throughout the duration of the radiation treatment course. Pretreatment planning CT scans with a median of 13 days before treatment (range: 2–27), mid‐treatment CT at 22 days after start of treatment (range: 13–38), and post‐treatment CT 65 days after start of treatment (range: 35–192). Patients received RT treatment to a total dose of 58–70 Gy, using daily 2.0–2.20 Gy, fractions for 30–35 fractions. The fan‐beam CT images were acquired using a Siemens 16‐slice CT scanner head protocol with 120 kV and current of 400 mAs. A helical scan with 1 rotation per second was used with a slice thickness of 2 mm and table increment of 1.2 mm. In addition to the imaging data, contours of anatomical structures for RT, demographic, and outcome measurements are provided.
Data format and usage notes
The dataset with DICOM files including images, RTSTRUCT files, and RTDOSE files can be found and publicly accessed in the Cancer Imaging Archive (TCIA, http://www.cancerimagingarchive.net/) as collection Head‐and‐neck squamous cell carcinoma patients with CT taken during pretreatment, mid‐treatment, and post‐treatment (HNSCC‐3DCT‐RT).
Discussion
This is the first dataset to date in TCIA which provides a collection of multiple CT imaging studies (pretreatment, mid‐treatment, and post‐treatment) throughout the treatment course. The dataset can serve a wide array of research projects including (but not limited to): quantitative imaging assessment, investigation on anatomical changes with treatment progress, dosimetry of target volumes and/or normal structures due to anatomical changes occurring during treatment, investigation of RT toxicity, and concurrent chemotherapy and RT effects on head‐and‐neck patients.
Keywords: CT, fan‐beam computed tomography, head‐and‐neck squamous cell carcinoma
1. Introduction
Cancer is a world‐wide disease affecting an estimated 2 million new cases and approximately 610,000 deaths every year. Head‐and‐neck cancers [mostly squamous cell carcinoma (HNSCC)] consist of about 8% of new cases and comprise approximately 7% of total cancer deaths. It is estimated that 134,000 new cases of HNSCC are diagnosed in the United States each year resulting in more than 40,000 deaths yearly.1
Approximately half of all cancer patients can benefit from radiation therapy (RT) in management of their disease.2 This is also true for head‐and‐neck (HN) cancer patients. HN patients in particular may undergo significant anatomical changes over a 6‐ to 7‐week course of radiation treatment.3, 4, 5 These anatomic changes can occur for a variety of reasons including decrease of tumor and lymph node volumes, weight loss due to poor oral intake, muscle mass alteration, fat distribution, and fluid shift in the body.3, 4, 6, 7 Gross changes in anatomy can be noticed with simple visual inspection of the patient, motion in the immobilization device, inability to align on cone‐beam CT, or more accurately with the acquisition of new computed tomography (CT) scans.6, 8 It is apparent that multiple CT images, acquired over the course of treatment, can facilitate the detection of these anatomy changes. Additionally, the use of CTs is fast, convenient, and it provides good contrast for visualization of these changes. Multiple CTs, however, contribute additional radiation dose to the patient, and as not being a standard of care, require additional efforts and resources. Nonetheless, close monitoring of visual changes (weight loss, loose immobilization, etc.), combined with CT imaging, can assist in treatment adaptation to potentially improve therapy outcome. To visualize the changes described above, HNSCC patient CTs were acquired at three time points — pre‐, mid‐, and post‐treatment. This is the first HNSCC dataset in The Cancer Imaging Archive (TCIA) with CT images acquired throughout the course of RT, which also includes results of normal tissue toxicity.
The HNSCC data available in TCIA include variety of Magnetic Resonance, Positron Emission Tomography (PET), and CT imaging studies, acquired “when available” (according to the data publication) as standard of care or when ordered.9, 10 The majority of that data include pre‐ and post‐treatment PET/CT scans which is valuable for evaluating and/or predicting tumor response,11, 12, 13, 14 but does not illustrate anatomy changes during therapy. At this time, limited MRI data (only nine subjects) for pretreatment, mid‐treatment, and post‐treatment HNSCC patients are available in TCIA, which however does not provide RT dose as well as supporting clinical information as opposed to our dataset.15 The data presented herein could be used alongside this MRI data to provide validation of their findings. Overall, our data hold new information not present in the TCIA datasets, since we include well‐annotated pre‐, mid‐, and post‐treatment CT images, as well as outcomes for tumor control and normal tissue toxicity of HNSCC patients after completion of therapy. Our dataset allows researchers to compare/validate existing clinical outcomes, visualize anatomical changes, potentially create improved treatment adaptation techniques, and test research ideas where imaging, tumor control, and normal tissue toxicity are studied.
The purpose of this work was to describe in detail longitudinal fan‐beam CT scans collected during pretreatment, mid‐treatment, and post‐treatment for 31 HNSCC patients and share this dataset with the community.
2. Acquisition and validation methods
2.A. Overview of the dataset
The dataset consists of three CT scans, contours, doses, clinical variables, and outcome measures from 31 HNSCC patients who underwent volumetric arc therapy (VMAT) at our institution from 2011 to 2017. VMAT was chosen due to superior dose conformity and tissue sparing.16, 17 All patients provided informed consent and were enrolled in an Institutional Review Board (IRB)‐approved protocol.
The study dates should be used only to discriminate and chronologically trace when the imaging studies were performed. Those study dates have been changed from the actual acquisition dates in order to protect patient confidentiality and comply with Health Insurance Portability and Accountability Act requirements. Modification was done in such a way as to preserve the relative time between acquisitions. Table 1 shows the modified dates for pretreatment, mid‐treatment, and post‐treatment, acquisition time to treatment start time (days), mid‐treatment fraction, and near‐post fraction.
Table 1.
Computed tomography imaging information for pretreatment, mid‐treatment, and post‐treatment. The data contain days from acquisition time to treatment start, days from treatment start to mid‐CT, and days from last treatment date to the post‐treatment scan. Number of slices per CT image for pre‐, mid‐, and post‐treatment are also included
| HN_P | Pre‐Tx (Days) | Mid‐Tx (Days) | Post‐Tx (Days) | Number of slices per CT image | ||
|---|---|---|---|---|---|---|
| Pre‐Tx | Mid‐Tx | Post‐Tx | ||||
| 001 | 14 | 33 | 47 | 196 | 209 | 198 |
| 002 | 12 | 21 | 35 | 193 | 186 | 196 |
| 003 | 7 | 17 | 37 | 174 | 204 | 229 |
| 004 | 24 | 22 | 36 | 235 | 226 | 207 |
| 005 | 27 | 17 | 37 | 207 | 206 | 217 |
| 007 | 12 | 22 | 36 | 220 | 217 | 158 |
| 008 | 12 | 18 | 38 | 207 | 189 | 197 |
| 009 | 10 | 38 | 52 | 191 | 200 | 197 |
| 010 | 13 | 23 | 38 | 191 | 185 | 194 |
| 011 | 7 | 23 | 38 | 200 | 206 | 205 |
| 012 | 14 | 23 | 36 | 178 | 168 | 193 |
| 013 | 11 | 24 | 44 | 180 | 182 | 201 |
| 014 | 20 | 15 | 37 | 215 | 189 | 223 |
| 015 | 5 | 23 | 49 | 222 | 216 | 227 |
| 016 | 13 | 23 | 38 | 181 | 199 | 209 |
| 017 | 8 | 29 | 41 | 189 | 183 | 198 |
| 018 | 12 | 20 | 41 | 209 | 192 | 196 |
| 019 | 2 | 31 | 107 | 180 | 179 | 172 |
| 020 | 14 | 23 | 90 | 225 | 182 | 175 |
| 022 | 13 | 30 | 102 | 181 | 168 | 188 |
| 023 | 12 | 25 | 86 | 206 | 179 | 192 |
| 024 | 19 | 24 | 79 | 179 | 221 | 219 |
| 026 | 13 | 28 | 97 | 194 | 171 | 165 |
| 027 | 13 | 20 | 94 | 219 | 228 | 218 |
| 028 | 14 | 21 | 91 | 228 | 178 | 190 |
| 029 | 12 | 16 | 86 | 244 | 236 | 195 |
| 030 | 14 | 16 | 105 | 196 | 230 | 218 |
| 032 | 14 | 14 | 56 | 216 | 221 | 205 |
| 033 | 12 | 23 | 91 | 193 | 221 | 193 |
| 034 | 8 | 21 | 68 | 192 | 206 | 185 |
| 035 | 13 | 13 | 192 | 202 | 234 | 213 |
Pretreatment planning CT scans were performed with a median of 13 days before treatment (range: 2–27), mid‐treatment CT was performed at 22 days after start of treatment (range: 13–38), and post‐treatment CT 65 days after start of treatment (range: 35–192). Patient imaging data were stored with identifiers in the form HN_PXXX, where XXX is a number from 001 to 035. Each number represents a different patient. Figure 1 shows pretreatment, mid‐treatment, and post‐treatment CT images with planning target volume (PTV) contours of two patients. Each contour color represents a different PTV delineated by the attending physicians. Table 1 shows the days when CT imaging was acquired prior to treatment start first column, after treatment start (second column), and after completion of therapy (third column). The other three columns denote the corresponding of images per CT scan (pretreatment, mid‐treatment, and post‐treatment).
Figure 1.
Pretreatment (a), Mid‐treatment (b), Post‐treatment (c) CT images with planning target volume contours of two patients (1 and 2). The different colors represent the same type of planning target volume across the two patients. [Color figure can be viewed at wileyonlinelibrary.com]
2.A.1. Inclusion/exclusion criteria
All patients were recruited according to the following criteria, specified in our IRB‐approved prospective imaging acquisition protocol. Inclusion: (a) patients had histologically confirmed HN tumors, (b) patients were treated with definitive RT or concurrent chemoradiation therapy, (c) patients treated with postoperative RT, for whom the gross tumor volume (GTV) or the resection cavity were visible on CT, such that it could be delineated as a target for RT, (d) patients were informed of the investigational nature of this study and were able to understand and to sign a written informed consent document. Exclusion: (a) pregnant or nursing women were not eligible to participate, (b) women of reproductive potential must have been offered a pretreatment pregnancy test and informed of the need to practice an effective contraceptive method, (c) patients younger than 18 years, (d) patients whose size and weight would not allow CT scanning, and (e) patients who did not finish the RT treatment course.
2.A.2. Patient demographics
Table 2 includes age, sex, tumor stage, tumor volume and location, prior surgery, performance status, and patient chemotherapy medication (if any). Performance status is a standard criteria used for measuring how the disease impacts patient's daily abilities. The Eastern Cooperative Oncology Group (ECOG) performance status is commonly used when planning trials to study a new treatment method. ECOG 0 indicates fully active, able to carry on all pre‐disease performance without restriction. ECOG 1 indicates restricted in physically strenuous activity but ambulatory and able to carry out work of a light or sedentary nature. ECOG 3 indicates capable of only limited self‐care, confined to bed or chair more than 50% of waking hours. ECOG 4 indicates completely disabled, cannot carry on any self‐care, totally confined to bed or char. ECOG 5 denotes patient's death.18 All the information in Table 2 was obtained from the institutional electronic medical record (EMR) system (Epic Systems Corporation, Verona, WI, USA) and its publication has been IRB approved.
Table 2.
Patient age, sex, tumor stage, tumor volume and location, prior surgery, performance status and chemotherapy medication
| HN_P | Age | Sex | Cancer staging | Tumor location | Prior surgery | Performance status | Chemotherapy medication |
|---|---|---|---|---|---|---|---|
| 001 | 64 | M | IVA | Left base of the tongue and retromolar trigone | Biopsy was obtained from the left retromolar mass as well as another biopsy from the posterior oral tongue | ECOG 0–1 | Cisplatin, Cetuximab |
| 002 | 83 | M | IVA | Right tonsil with cervical adenopathy | Cataract surgery (both eyes), corneal transplant (left eye), FNA biopsy of the bulky cervical adenopathy on the right side | ECOG 0–1 | Carboplatin, Cetuximab |
| 003 | 87 | M | IVA | Left base of the tongue | Carpal tunnel release, elbow surgery, knee surgery, tonsillectomy and adenoidectomy, tongue biopsy | ECOG 0–1 | Carboplatin, Cetuximab |
| 004 | 66 | M | IVA | Left base of the tongue | Total hip arthroplasty | ECOG 0–1 | Erbitux |
| 005 | 55 | M | IVA | Base of tongue | Hemorrhoid surgery | ECOG 0–1 | Cisplatin |
| 007 | 52 | M | IVA | Left tonsil | Tonsil biopsy | ECOG 0–2 | Cisplatin |
| 008 | 71 | M | III | Right tonsil with ipsilateral nodal disease. | Appendectomy, biopsy tonsil | ECOG 0–1 | Cisplatin |
| 009 | 59 | M | IVA | Anterior floor of the mouth | Cholecystectomy, and history of thoracic surgery, tumor removal at floor of the mouth | ECOG 0–1 | Cisplatin |
| 010 | 59 | F | III | Left lateral border of the tongue | Cesarean section, appendectomy, cholecystectomy, tongue surgery, lymph removal at neck | None | None |
| 011 | 87 | M | II | Left lateral oral tongue | Coronary angioplasty, Cataract removal (right), Prostatectomy, colonoscopy with or without biopsy | None | None |
| 012 | 55 | F | IIB | Right oral tongue | Right hemiglossectomy, right revision radical neck dissection, radial forearm free flap, tracheotomy, hysterectomy, tonsillectomy, laryngoscopy/bronchoscopy/esophagoscopy, glossectomy, tongue surgery | None | Cisplatin |
| 013 | 45 | M | IVA | Retromolar trigone | Neck surgery (retromolar trigone) | ECOG 0 | Cisplatin |
| 014 | 69 | M | IVA | Left mandible | Tongue biopsy, tonsillectomy | None | Cisplatin |
| 015 | 71 | M | IVA | Lymph nodes of head, face, and neck | Umbilical hernia repair, Surgery of lip, cosmetic surgery, dissection neck | ECOG 0–1 | Cisplatin |
| 016 | 80 | M | IVA | Left base of the tongue, both tonsils with extensive lymph node | Laryngectomy, inguinal hernia repair | ECOG 1 | Cetuximab |
| 017 | 84 | M | IIA | Base of tongue | Appendectomy, hernia repair, prostate surgery, cardiac catheterization, cataract surgery | ECOG 0–3 | None |
| 018 | 67 | M | IVA | Larynx | Tonsillectomy and the vocal cord biopsy, total laryngectomy, bilateral neck dissection, tracheoesophageal puncture | ECOG 0 | None |
| 019 | 70 | F | IVA | Left parotid gland | Appendectomy, cholecystectomy, sinus surgery, and neck and ear surgery | None | None |
| 020 | 59 | M | III | Larynx (right true vocal cord) | Hemorrhoid surgery, colonoscopy | None | Cisplatin |
| 022 | 43 | F | III | Right tongue base | Tonsillectomy | ECOG 1 | Cisplatin |
| 023 | 71 | M | III | Nasal cavity | Appendectomy, hernia repair, nasal endoscopy, and sinus surgery | None | None |
| 024 | 66 | M | IVA | Oropharynx | Vulva removal, knee surgery, myringotomy (right), tonsillectomy | None | Cisplatin (with Mannitol). |
| 026 | 61 | F | IVA | Larynx | Endoscopy, biopsy, total laryngectomy, cholecystectomy | ECOG 1–2 | Cisplatin |
| 027 | 65 | M | III | Larynx‐ right TVC | Cholecystectomy, IR US biopsy thyroid (right) | ECOG 1 | Carboplatin |
| 028 | 68 | M | III | Occipital scalp | Tumor excision in head, colonoscopy w/or w/o biopsy, upper gastrointestinal endoscopy, resect tumor soft tissue neck/anterior thorax, removal nodes, neck, cervical, tunneled venous port placement | None | Cisplatin |
| 029 | 64 | M | I | Right tonsil | Coronary angioplasty with stent placement, aorto‐femoral bypass graft, cardiac defibrillator placement, pacemaker placement | None | None |
| 030 | 80 | M | IVA | Left hard palate | Cryoablation prostate transperineal, tonsillectomy, repair hernia inguinal | None | None |
| 032 | 51 | M | III | Paranasal sinus | Cardiac surgery, coronary angioplasty with stent placement, sinus surgery | None | None |
| 033 | 45 | F | IIA | Oral tongue | Tubal ligation | None | None |
| 034 | 56 | F | III | Nasal cavity | Cesarean section, cholecystectomy, nasal polyp removal, nasal endoscopy, removal part ethmoid, maxillary sinusotomy, upper jaw maxillectomy | None | None |
| 035 | 41 | F | I | Right true vocal cord, glottic larynx | larynx surgery, cesarean section | None | None |
2.B. Acquisition
2.B.1. CT
In this work CT scan refers to a series of images, while an image represents a single frame of the CT scan. Longitudinal fan‐beam CT scans were acquired using a Siemens 16‐slice CT scanner with standard clinical HN protocol. The protocol utilized 120 kV potential and tube current of 400 mAs. A helical scan with 1 rotation per second was used with a slice thickness of 2 mm and table increment of 1.2 mm. The field of view was 50 × 50 cm, mapped to a slices of 512 × 512 pixels, with a pixel size of 0.0977 × 0.0977 cm2.
All CT scans were acquired on the same scanner, and a Hounsfield Unit‐to‐electron density calibration table reproduced from the clinical treatment planning system is listed in Table 3 for reference.
Table 3.
Hounsfield Unit‐to‐electron density conversion table
| CT number | Density |
|---|---|
| 0 | 0.000 |
| 179 | 0.190 |
| 499 | 0.459 |
| 936 | 0.952 |
| 976 | 0.976 |
| 999 | 1.000 |
| 1047 | 1.043 |
| 1054 | 1.052 |
| 1223 | 1.117 |
| 1900 | 1.512 |
| 5000 | 12.000 |
2.B.2. Segmentation
In addition to the imaging data, the dataset contains contours of anatomical structures for RT. The contours were outlined by the attending radiation oncologists for treatment planning purposes. The RT treatment plans were generated with Eclipse® (Varian Medical Systems, Inc., Palo Alto, CA, USA) treatment planning system. Table 4 shows the organs at risk (OARs), where X denotes if the contour was present in the plan (different patients had different OARs outlined, depending on tumor location and clinician discretion). OARs include spinal cord, right parotid gland, left parotid gland, brainstem, esophagus, larynx, and mandible. Table 5 includes the planning target volume (PTV) doses. OARs and PTV were included together because they are delineated by the physician prior to treatment planning and are used in the treatment plan optimization. Patients had several levels of prescription, and therefore several PTVs were outlined. The corresponding doses are listed in Table 5. The clinical target volume to PTV contour expansions were generated on per‐patient basis, depending on the individual therapeutic needs and clinician's decision. Figure 2 shows the OARs described above for three different patients for pretreatment CTs in axial, sagittal, and coronal views.
Table 4.
Contoured structures organs at risk per patient in Original, Mid, and Post Images
| HN_P | Cord | Rparotid | Lparotid | Brainstem | Esophagus | Larynx (GSL) | Mandible | |
|---|---|---|---|---|---|---|---|---|
| Original | 001 | x | x | x | x | x | ||
| Mid | x | x | x | |||||
| Near Post | x | x | x | x | x | |||
| Original | 002 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | |||||
| Original | 003 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 004 | x | x | x | x | x | x | x |
| Mid | x | x | x | x | x | x | x | |
| Near Post | x | x | x | x | x | x | x | |
| Original | 005 | x | x | x | x | x | x | x |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 007 | x | x | x | x | x | x | x |
| Mid | x | x | x | x | x | x | x | |
| Near Post | x | x | x | x | x | x | x | |
| Original | 008 | x | x | x | x | x | ||
| Mid | x | x | x | x | x | |||
| Near Post | x | x | x | x | x | |||
| Original | 009 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 010 | x | x | x | x | x | ||
| Mid | x | x | x | x | x | |||
| Near Post | x | x | x | x | x | |||
| Original | 011 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 012 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 013 | x | x | x | x | x | ||
| Mid | x | x | x | x | x | |||
| Near Post | x | x | x | x | x | |||
| Original | 014 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 015 | x | x | x | x | x | ||
| Mid | x | x | x | x | x | |||
| Near Post | x | x | x | x | x | |||
| Original | 016 | x | x | x | x | x | ||
| Mid | x | x | x | x | x | |||
| Near Post | x | x | x | x | x | |||
| Original | 017 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 018 | x | x | x | x | x | ||
| Mid | x | x | x | x | x | |||
| Near Post | x | x | x | x | x | |||
| Original | 019 | x | x | x | x | x | ||
| Mid | x | x | x | x | x | |||
| Near Post | x | x | x | x | x | |||
| Original | 020 | x | x | x | x | x | ||
| Mid | x | x | x | x | x | |||
| Near Post | x | x | x | x | x | |||
| Original | 022 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 023 | x | x | x | x | ? | ||
| Mid | x | x | x | x | ? | |||
| Near Post | x | x | x | x | ? | |||
| Original | 024 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 026 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 027 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | |||
| Original | 028 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 029 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 030 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 032 | x | x | x | x | x | ||
| Mid | x | x | x | x | x | |||
| Near Post | x | x | x | x | x | |||
| Original | 033 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 034 | x | x | x | x | x | x | |
| Mid | x | x | x | x | x | x | ||
| Near Post | x | x | x | x | x | x | ||
| Original | 035 | x | ||||||
| Mid | x | |||||||
| Near Post | x |
Table 5.
Planning target volumes in Gy per patient
| HN_P | PTV1 | PTV2 | PTV3 | PTV4 |
|---|---|---|---|---|
| 001 | 60 | 54 | ||
| 002 | 70 | 60 | 54 | |
| 003 | 60 | 54 | ||
| 004 | 70 | 60 | 54 | |
| 005 | 70 | 56 | 52.5 | |
| 007 | 70 | 60 | 54 | |
| 008 | 70 | 60 | 54 | |
| 009 | 64.5 | 60 | 54 | |
| 010 | 60 | 54 | ||
| 011 | 60 | 54 | ||
| 012 | 66 | 60 | 54 | |
| 013 | 66 | 54 | ||
| 014 | 66 | 60 | 54 | |
| 015 | 66 | 60 | 54 | |
| 016 | 66 | 54 | ||
| 017 | 70 | 66 | 63 | 56 |
| 018 | 60 | 54 | ||
| 019 | 60 | 54 | ||
| 020 | 70 | 60 | 54 | |
| 022 | 70 | 60 | 54 | |
| 023 | 60 | |||
| 024 | 70 | 60 | 54 | |
| 026 | 66 | 60 | 54 | |
| 027 | 70 | 60 | 54 | |
| 028 | 60 | 54 | ||
| 029 | 63 | |||
| 030 | 60 | 54 | ||
| 032 | 66 | 60 | ||
| 033 | 60 | 54 | ||
| 034 | 60 | |||
| 035 | 66 |
Figure 2.
Organs at risk contours for three different patients outlined on the pretreatment CT images in axial, sagittal, and coronal views. Each organ at risk is represented in a different color. The different colors represent the same organ at risk across the three patients. [Color figure can be viewed at wileyonlinelibrary.com]
2.B.3. Clinical parameters
Table 6 lists clinical treatment information for all subjects including prescription, dose per fraction, total number of fractions, PTV volume, VMAT arcs, and energy. Patients underwent RT treatment to a total dose of 58–70 Gy, using daily 2–2.20 Gy, fractions for 30–35 fractions. The reported PTV volumes are combination of all PTVs shown in Table 5, since they represent the total treated volume.
Table 6.
Treatment information: Total dose, number of fractions, dose per fraction, number of arcs (full arcs are a complete rotation while partial arcs are denoted by degrees of rotation), and beam energy
| HN_P | Total dose (Gy) | Number of fx | Dose/fx (Gy) | PTV vol (cm3) | VMAT arcs | Energy (MV) |
|---|---|---|---|---|---|---|
| 001 | 69.96 | 33 | 2.12 | 650.37 | 2 full | 6 |
| 002 | 69.96 | 33 | 2.12 | 561.63 | 2 full | 6 |
| 003 | 69.96 | 33 | 2.12 | 471.64 | 2 full | 6 |
| 004 | 70 | 35 | 2.00 | 744.30 | 2 full | 6 |
| 005 | 70 | 35 | 2.00 | 481.63 | 2 full | 6 |
| 007 | 69.96 | 33 | 2.12 | 629.86 | 2 full | 6 |
| 008 | 69.96 | 33 | 2.12 | 573.82 | 2 full | 6 |
| 009 | 65 | 30 | 2.15 | 604.77 | 2 full | 6 |
| 010 | 60 | 30 | 2.00 | 351.96 | 2 full | 6 |
| 011 | 60 | 30 | 2.00 | 581.63 | 2 full | 6 |
| 012 | 66 | 30 | 2.20 | 524.45 | 2 full | 6 |
| 013 | 66 | 33 | 2.00 | 816.94 | 2 full, 1 arc :330–30a | 6 |
| 014 | 66 | 30 | 2.20 | 602.99 | 2 full | 6 |
| 015 | 66 | 33 | 2.00 | 602.29 | 2 full | 6 |
| 016 | 66 | 33 | 2.00 | 486.86 | 2 full | 6 |
| 017 | 70 | 35 | 2.00 | 724.45 | 2 full | 6 |
| 018 | 60 | 30 | 2.00 | 595.22 | 2 full | 6 |
| 019 | 60 | 30 | 2.00 | 267.13 | 3 full | 6 |
| 020 | 69.96 | 33 | 2.12 | 440.64 | 2 full | 6 |
| 022 | 69.96 | 33 | 2.12 | 447.25 | 3 full | 6 |
| 023 | 60 | 30 | 2.00 | 170.57 | 3 arcs: 208–152a, 1arc: 320–20a | 6 |
| 024 | 69.96 | 33 | 2.12 | 538.83 | 3 full | 6 |
| 026 | 66 | 33 | 2.00 | 676.72 | 3 full | 6 |
| 027 | 69.96 | 33 | 2.12 | 668.63 | 3 full | 6 |
| 028 | 60 | 30 | 2.00 | 537.14 | 3 full | 6 |
| 029 | 63.6 | 30 | 2.12 | 63.22 | 2 full | 6 |
| 030 | 60 | 30 | 2.00 | 360.14 | 3 arcs: 200–160a | 6 |
| 032 | 60 | 30 | 2.00 | 139.24 | 3 full | 6 |
| 033 | 60 | 30 | 2.00 | 494.93 | 3 full | 6 |
| 034 | 60 | 30 | 2.00 | 140.96 | 3 full | 6 |
| 035 | 66 | 33 | 2.00 | 42.56 | 2 full | 6 |
IEC coordinate system.
2.B.4. Outcome information
Table 7 includes treatment related toxicities for parotid glands and oral cavity. Table 8 includes patient weight change during the course of treatment and the net loss after the end of therapy. Grading for toxicity was measured using Common Terminology Criteria for Adverse events version 4. Most common toxicities were xerostomia and mucositis, which appear in our database on average from day 22 to 186 days post radiotherapy. All this information was obtained from the institutional EMR system and its publication has been approved by the IRB.
Table 7.
Patient toxicities associated with the pretreatment, mid‐treatment, and post‐treatment time points
| HN_P | Outcome mid‐treatment (tx) | Outcome near end treatment (tx) | Outcome 3–6 months post |
|---|---|---|---|
| 001 | Grade 2 xerostomia, Grade 3 mucositis | Grade 3 mucositis involving the Left oral tongue, facial arch | No palpable adenopathy of the head, neck or supraclavicular regions. Oral cavity and oropharynx are without suspicious lesions. Palpation of the oral tongue, floor of mouth, tongue base and tonsils bilaterally is negative. Significant mucositis and edema, mild xerostomia |
| 002 | Grade 2 mucositis involving the soft palate, Grade 1 mucositis involving the bilateral lateral tongue | Grade 3 mucositis involving the right oral tongue laterally and Grade 2 involving the left soft palate and left oral tongue | No masses are seen in the upper aerodigestive tract. Mucosal lining of the pharynx and larynx shows post‐treatment changes. Superficial small ulcers on the posterior pharyngeal wall without any obviously suspicious lesions. Both vocal cords have normal aspect and range of motion, airway is patent. Significant xerostomia, odynophagia |
| 003 | Grade 1 xerostomia, Grade 1 mucositis of the left buccal mucosa and left lower gingiva | Grade 2 mucositis in the left lateral aspect of the base of tongue, Grade 3 mucositis on the soft palate | Nasal cavity, nasopharynx, oropharynx, supraglottic laryngeal structures, pyriform sinuses, and glottic larynx were without suspicious lesions. Vocal cords were mobile and met in midline. No edema. Voice was normal. Xerostomia present |
| 004 | Grade 1 oral mucositis | Grade 2 oral mucositis | Nasal cavity, nasopharynx, oropharynx, supraglottic laryngeal structures, pyriform sinuses, and glottic larynx were without suspicious lesions. Vocal cords were mobile and met in midline. Minimal bilateral arytenoid edema. Voice was normal. Xerostomia present |
| 005 | No toxicity | Grade 2 dermatitis over the treated area, Grade 1 mucositis over the left posterior pharyngeal wall | Nasal cavity, nasopharynx, oropharynx, supraglottic laryngeal structures, pyriform sinuses, and glottic larynx were without suspicious lesions. Vocal cords were mobile and met in midline. No edema. Voice was normal |
| 007 | Grade 3 mucositis, Grade 2 mucositis over the soft palate and posterior oropharyngeal wall | Grade 3 mucositis over the soft palate and posterior oropharyngeal wall, with areas of Grade 2 along the lateral aspects of the tongue | Presents with dysphagia related to reduced base of tongue retraction. Presents with modest narrowing at the cricopharyngeal area that reduces bolus flow at this level. Persistent mucositis accompanied by taste change, dry mouth, reduced appetite and weight loss |
| 008 | Grade 1 mucositis over the hard and soft palate | Grade 3 mucositis of soft palate and tongue | No lesions or suspicious masses in the upper airway. Both vocal cords were mobile bilaterally and the airway was adequate. Nasal cavity, nasopharynx, oropharynx, supraglottic laryngeal structures, pyriform sinuses, and glottic larynx were without suspicious lesions. Significant blunting of the epiglottis. Vocal cords were mobile and met in midline. No edema. Voice was normal. Xerostomia present |
| 009 | No toxicity | Grade 3 mucositis in the floor of mouth, with Grade 2 in the tonsillar pillars | Significant for mild‐to‐moderate degree of mucositis and mouth sores as a postradiation change. Moderate degree of dry xerostomia. No lesions or suspicious masses in the upper airway. Both vocal cords were mobile bilaterally and the airway was adequate |
| 010 | Grade 1 mucositis of the oral cavity except the left lateral tongue which is Grade 2, Grade 1 xerostomia | Grade 3 confluent mucositis of the left oropharynx and left lateral tongue | No palpable adenopathy of the head, neck, or supraclavicular regions. Oral cavity and oropharynx are without suspicious lesions. Nasal cavity, nasopharynx, oropharynx, supraglottic laryngeal structures, pyriform sinuses, and glottic larynx were without suspicious lesions. Vocal cords were mobile and met in midline. No edema. Voice was normal |
| 011 | Grade 3 mucositis in mouth (buccal mucosa, vestibule of mouth left tongue) | Grade 3 mucositis in mouth (buccal mucosa, vestibule of mouth left tongue) | Dry mouth as the only side effect. Oral cavity, oropharynx, hypopharynx, larynx, nasopharynx, nose and necks, including the use of a fiber optic scope shows no evidence of recurrence |
| 012 | Grade 2 mucositis | Grade 3 mucositis in oral cavity is limited to the left side | CT of the neck failed to show tumor recurrence and CT of the chest was normal. Rash noted. No masses are palpable in the neck |
| 013 | No toxicity | Grade 2 mucositis on the right ventrolateral aspect of the tongue and on the inner lip, Grade 2 mucositis on the right ventrolateral aspect of the tongue and on the inner lip | Developed a rapid recurrence in the right temporal region on the margin of the radiation field and completed chemo and re‐irradiation. Increased thickening of the epiglottis and supraglottic larynx, could be related to post‐treatment changes. No palpable adenopathy of the head, neck, or supraclavicular regions. Oral cavity and oropharynx are without suspicious lesions. Moderate thrush in the oral cavity. Mild xerostomia |
| 014 | Grade 1/2 mucositis limited to the oropharynx | Grade 2 mucositis of the soft palate | No toxicity |
| 015 | Grade 2 mucositis at underside of tongue and mucosal aspect of the lower lip | Grade 2 mucositis over the lip and tongue | Lymph nodes in mediastinum, lung nodule, and hypermetabolic upper abdominal preaortic lymphadenopathy 5. Mild hepatomegaly and evidence of splenomegaly. Transbronchial biopsy of right and left hilar lymph nodes and both positive for cancer |
| 016 | Grade 1 mucositis of oral cavity | Grade 2 mucositis | Complete stenosis of the distal hypopharynx. Oral cavity, oropharynx, nasal cavity, nasopharynx, new hypopharynx, neck, and stoma are without evidence of recurrence. Dysphagia — receiving PEG tube feedings, but likely getting too little calories based on nutritional assessment. Weight loss, intermittent diarrhea |
| 017 | Grade 1 mucositis | Grade 2 xerostomia, Grade 2 mucositis of the right roof of mouth | Nasal cavity, nasopharynx, oropharynx, supraglottic laryngeal structures, pyriform sinuses, and glottic larynx were without suspicious lesions. Vocal cords were mobile and met in midline. No edema. Voice was normal. Grade 2 xerostomia |
| 018 | No toxicity | No toxicity | Mucosal excoriations and sloughing extending circumferentially to about 2–3 cm from the stomal opening. No extensive crusting or bleeding. Trachea to the carina within normal. Both vocal cords were mobile bilaterally and the airway was adequate. Later examination revealed significant edema of mucosa in the nasopharynx |
| 019 | Grade 3 mucositis of left posterior buccal mucosa | No toxicity | Left parotid adenocarcinoma post‐surgery and radiation. Left mixed hearing loss and tinnitus. Left facial/auricular discomfort — likely left TMJ dysfunction (TMJ capsule removed from tumor involvement) |
| 020 | No toxicity | No toxicity | Occasional odynophagia when swallows. Minimal xerostomia. No masses are seen in the upper aerodigestive tract. Mucosal lining of the pharynx and larynx is normal except for post‐treatment changes. Both vocal cords have relatively normal aspect and the right cord has slightly reduced range of motion. Airway is patent. No evidence of persistent disease and no evidence of progression of chondronecrosis |
| 022 | Grade 2 mucositis involving the right soft palate, right lateral oral tongue and right buccal surface, Grade 2 xerostomia | No toxicity | Nasal cavity, nasopharynx, oropharynx, supraglottic laryngeal structures, pyriform sinuses, and glottic larynx were without suspicious lesions. Blunting of the epiglottis and bilateral arytenoid edema. Vocal cords were mobile and met in midline. Voice was normal |
| 023 | No toxicity. | No toxicity. | Nasal crusts at floor of nasal cavity and lateral nasal wall removed. Sinuses all patent. Few granulations suctioned away central skull base with normal ASB otherwise. |
| 024 | Grade 2 xerostomia, Grade 1 mucositis involving the soft palate and lateral tongue | Grade 2 xerostomia | No adenopathy in the neck. Nasal cavity, nasopharynx, oropharynx, supraglottic laryngeal structures, pyriform sinuses, and glottic larynx were without suspicious lesions. Left arytenoid edema was noted. Vocal cords were mobile and met in midline. No edema. Voice was normal |
| 026 | No toxicity | No toxicity | Airway is patent. Mucosa shows post‐treatment changes. Cavities are widely patent and clean including maxillary, frontal, and sphenoid sinuses. Skull base is covered with relatively normal mucosa and the right inferior turbinate flap appears healthy. No lesions, polyps, or secretions are seen |
| 027 | Grade 1 mucositis posteriorly in pharynx, Grade 2 mucositis of right lateral tongue, Grade 3 laryngeal mucositis | Grade 1 mucositis posteriorly in pharynx, Grade 2 mucositis of right lateral tongue, Grade 3 laryngeal mucositis | Nasal cavities, nasopharynx, oropharynx, hypopharynx, and larynx shows no abnormality. Both vocal cords are fully mobile. Pyriform sinuses, post cricoid area, and vallecula are normal. No lesions seen |
| 028 | No toxicity | No toxicity | Large occipital mass excised from the occipital scalp was a squamous cell carcinoma by an outside surgeon. A new lesion was seen near the original site. He underwent a left posterior neck dissection, levels 2 through 4, and level 5 with adjacent tissue transfer and trapezius flap |
| 029 | Grade 2 mucositis of the right retrimolar trigone and anterior tonsillar pillar, and right lateral tongue base | No toxicity | Patient complains about pain in the throat, neck, and posterior cervical spine for which he is requesting pain medications. Nasal cavity, nasopharynx, oropharynx, supraglottic laryngeal structures, pyriform sinuses, and glottic larynx were without suspicious lesions. Vocal cords were mobile and met in midline. No edema. Voice was normal |
| 030 | No toxicity | Grade 1 xerostomia | No toxicity |
| 032 | No toxicity | No toxicity | Heavy growth of Staphylococcus aureus. Crusting after radiation — stable amount |
| 033 | Grade 2 mucositis in the flap and the tongue, Grade 2 xerostomia | Grade 1 mucositis | No toxicity |
| 034 | No toxicity | Grade 1 mucositis, Grade 2 mucositis of right nares | Right side has normal mucosa. Left side has mild contracture. No lesions seen. Ethmoid cavity normal. Crusts filling mega antrostomy cavity. Patchy area of exposed bone in maxillary sinus. Nasal secretions discolored secretions left. Nasal septum mild right deviation and no perforation appreciated, synechia formation between the inferior turbines and septum |
| 035 | No toxicity | No toxicity | Supraglottic structures demonstrate mild postradiation edema. The airway is widely patent, and nasal cavity, nasopharynx, oropharynx, supraglottic laryngeal structures, pyriform sinuses, and glottic larynx were without suspicious lesions. Vocal cords were mobile and met in midline. No edema. Voice slightly raspy |
Table 8.
Patient weight change during radiotherapy, and net weight loss (lbs) after treatment completion
| HN_P | Pre‐tx weight | Mid‐tx weight | Post‐tx weight | Net weight loss |
|---|---|---|---|---|
| 001 | 138 | 138 | 131 | 7 |
| 002 | 164 | 164 | 157 | 7 |
| 003 | 210 | 210 | 195 | 15 |
| 004 | 201 | 201 | 194 | 7 |
| 005 | 187 | 187 | 181 | 6 |
| 007 | 216 | 216 | 207 | 9 |
| 008 | 159 | 159 | 148 | 11 |
| 009 | 172 | 172 | 163 | 9 |
| 010 | 155 | 155 | 145 | 10 |
| 011 | 188 | 188 | 188 | 0 |
| 012 | 137 | 137 | 127 | 10 |
| 013 | 214 | 214 | 207 | 7 |
| 014 | 141 | 141 | 125 | 16 |
| 015 | 227 | 227 | 218 | 9 |
| 016 | 84 | 84 | 84 | 0 |
| 017 | 146 | 146 | 138 | 8 |
| 018 | 150 | 150 | 148 | 2 |
| 019 | 124 | 124 | 118 | 6 |
| 020 | 123 | 123 | 113 | 10 |
| 022 | 163 | 163 | 132 | 31 |
| 023 | 165 | 165 | 168 | −3 |
| 024 | 175 | 175 | 172 | 3 |
| 026 | 147 | 147 | 128 | 19 |
| 027 | 205 | 205 | 170 | 35 |
| 028 | 200 | 200 | 180 | 20 |
| 029 | 134 | 134 | 138 | −4 |
| 030 | 161 | 161 | 143 | 18 |
| 032 | 199 | 199 | 196 | 3 |
| 033 | 162 | 162 | 158 | 4 |
| 034 | 188 | 188 | 188 | 0 |
| 035 | 149 | 146 | 147 | 2 |
2.B.5. Known issues and limitations
There are a few known issues and limitations of this dataset. We have a limited number of HNSCC patients (31), therefore the conclusions derived from this dataset may not be representative of the general HNSCC population. Another limitation is the differences of the outlined anatomical structures across patients. All the structures were outlined at the discretion of the attending physicians and since different patient had different tumors, some of the OARs were not included for every patient (Table 4 indicates which structures are present for each patient). In addition, PTV prescription doses and fractionation schemes are tailored to each individual and therefore they are not uniform across patients, as presented in Table 5.
2.C. Data validation
As all the patients in this dataset were treated at our institution and images were acquired using the same CT scanner by licensed radiation therapists. The utilized reconstruction algorithm was consistent for all scans in the dataset. The transfer of the images to the treatment planning system followed the same institutional protocol. After image import in the treatment planning system, all images were reviewed by licensed dosimetrists and/or physicist for any artifact prior to performing any segmentation and treatment planning according to the guidelines from AAPM Task Group 66.19 CT quality assurance tests for mechanical and imaging performance were regularly performed on daily, monthly, and annual basis, thereby ensuring adequate, reliable, and reproducible SC scanner performance.
In order to verify the dataset, the DICOM files with the corresponding RTSTRUCT and RTDOSE were cross‐checked with the original (non‐anonymized) data. All patient data were then transferred from MIM (MIMCorp, Cleveland, OH, USA) to Eclipse after anonymization to ensure compatibility with the treatment planning system.
3. Data format and usage notes
Data were exported from Eclipse in DICOM (images), DICOM‐RTSTRUCT (structures), and DICOM‐RTDOSE (doses) format and then imported into MIM. It should be noted that we have two DICOM‐RTst per imaging set, one with series description MIM structures and another named ARIA Rad Onc Structure Sets. The RTst‐MIM structures was renamed by us to remain consistent with Table 5 structures. On the other hand, RTst‐ARIA Rad Onc Structure Sets were the original structures as part of treatment planning and delivery. Our recommendation is to use RTst‐MIM structures so that the information on Table 5 is helpful. Data were then imported into TCIA at the National Institutes of Health for storage, where it was additionally post‐processed to ensure protection of potentially confidential information.20
HN_PXXX is consistent among all tables and TCIA data. The PatientID DICOM tag has a similar identification. Images can be identified by the DICOM study date, as mentioned previously, but the study dates are offset to protect identification while remain consistent with relative to time between scans. DICOM series description contains patient identifier, patient ID, and study ID. The patient name is HN_PXXX, while patient ID and study ID are ANONXXX. ANONXXX was a randomized name created by MIM software to de‐identify the patients. DICOM tags related to acquisition such as the technique, image orientation, and spacing were preserved during de‐identification and anonymization.
The dataset of HNSCC patients with multiple CT acquired during the course of therapy can be found at the TCIA website http://www.cancerimagingarchive.net/. Alternatively, the dataset can be accessed by digital object identifier (DOI) https://doi.org/10.7937/k9/tcia.2018.13upr2xf.
The total size of the dataset is 14.4 GB, with 372 individual DICOM files consisting of 93 images containing 18906 individual slices, 186 RTSTRUCT files, and 93 RTDOSE files. DICOM and DICOM‐RT compatibility have been tested with Eclipse and MIM (MIM Software, Cleveland, OH, USA).
The DOI link above points directly to the dataset. To download the data, the user needs to select the download button under Data Access. A TCIA Download Manager will appear where a directory to save the data can be selected. Once downloaded, the data will appear in separate folders titled HN_PXXX, representing different patients. Each folder contains the pretreatment, mid‐treatment, and post‐treatment folders, and within each of these are imaging data, a RTSTRUCT, and RTDOSE. In order to view these data, a DICOM accessible software such as standard RT treatment planning system, MIM, Velocity, or an alternative DICOM viewer (such as DicomBrowser at http://nrg.wustl.edu/software/dicom-browser) will be required.
4. Discussion
This dataset included imaging information, contours of tumors and normal tissues, demographics, and outcomes for tumor control and normal tissue complications. This information will allow researchers to measure gross tumor changes, validate their existing data, and potentially create new measures to treatment adaptation. In the future, this dataset can be used for contouring purposes where automated or semiautomated algorithms can be trained or validated on the data. Furthermore, the data can be used in radiomics studies for predictions of tumor control and/or normal tissue complications. The dataset can also be utilized for training and/or validation in delta‐radiomics studies. Another venue for application of this data might be in proton studies, where plan robustness can be tested on changing patient anatomy.
The HNSCC data available in TCIA include a variety of MRI, PET, and CT imaging studies with unclear dates and timelines creating difficulty in seeing anatomical changes during radiation treatment and success of such treatment. Majority of the available data only include pretreatment and post‐treatment PET/CT scans which does not illustrate anatomy changes during treatment. Our dataset adds to the collective HNSCC data in TCIA and it can be used in conjunction with pretreatment and post‐treatment PET/CT and MRI for pretreatment, mid‐treatment, and post‐treatment HNSCC data found in the repository.
5. Conclusion
The purpose of this manuscript was to describe in detail longitudinal fan‐beam CT scans collected during pre‐, mid‐, and post‐treatment for 31 HNSCC patients and to share these data with the community. Despite being relatively small, this dataset adds more information to the already existing imaging archive in the TCIA, and thereby contributes to the Big Data paradigm.
Acknowledgments
We thank UM radiation therapists who performed the imaging and the RT. The research was funded by the NIH grant R01CA163370. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
References
- 1. American Cancer Society . Cancer Facts & Figures 2018. Atlanta, GA: American Cancer Society; 2018. [Google Scholar]
- 2. Gelband H, Jha F, Sankaranarayanan R, Horton S. Cancer: Disease Control Priorities, Third Edition (Volume 3). Washington, DC: The International Bank for Reconstruction and Development/The World Bank; 2015. Chapter 14. [PubMed] [Google Scholar]
- 3. Barker JL, Garden AS, Ang KK, et al. Quantification of volumetric and geometric changes occurring during fractionated radiotherapy for head‐and‐neck cancer using an integrated CT/linear accelerator system. Int J Radiat Oncol Biol Phys. 2004;59:960–970. [DOI] [PubMed] [Google Scholar]
- 4. Robar JL, Day A, Clancey J, et al. Spatial and dosimetric variability of organs at risk in head‐and‐neck intensity‐modulated radiotherapy. Int J Radiat Oncol Biol Phys. 2007;68:1121–1130. [DOI] [PubMed] [Google Scholar]
- 5. Lee C, Langen K, Kupelian P, et al. Positional and volumetric changes in parotid glands during head‐and‐neck radiation therapy assessed using deformable image registration. Int J Radiat Oncol Biol Phys. 2007;69:S203–S204. [Google Scholar]
- 6. Lee C, Langen KM, Lu W, et al. Assessment of parotid gland dose changes during head‐and‐neck cancer radiotherapy using daily megavoltage computed tomography and deformable image registration. Int J Radiat Oncol Biol Phys. 2008;71:1563–1571. [DOI] [PubMed] [Google Scholar]
- 7. Perni S, Mohamed AS, Garden AS, et al. Pretreatment computed tomography–based volumetric tumor growth velocity: a novel prognostic indicator for head‐and‐neck cancer. Int J Radiat Oncol Biol Phys. 2016;96:E341–E342. [Google Scholar]
- 8. Rali A, Anglesio SM, Mortellaro G, et al. Volumetric and positional changes of planning target volumes and organs at risk using computed tomography imaging during intensity‐modulated radiation therapy for head–neck cancer: an ‘Old’ adaptive radiation therapy approach. Radiol Med (Torino). 2014;119:714–720. [DOI] [PubMed] [Google Scholar]
- 9. National Cancer Institute Clinical Proteomic Tumor Analysis Consortium (CPTAC) . (2018). Radiology Data from the Clinical Proteomic Tumor Analysis Consortium Head and Neck Squamous Cell Carcinoma [CPTAC‐HNSCC] Collection. The Cancer Imaging Archive. 10.7937/k9/tcia.2018.uw45nh81 [DOI]
- 10. Zuley ML, Jarosz R, Kirk S, et al. (2016). Radiology data from the cancer genome atlas head‐neck squamous cell carcinoma [TCGA‐HNSC] collection. The Cancer Imaging Archive. 10.7937/k9/tcia.2016.lxkq47ms [DOI]
- 11. Vallières M, Kay‐Rivest E, Perrin LJ, et al. (2017). Data From Head‐Neck‐PET‐CT. The cancer imaging archive. 10.7937/k9/tcia.2017.8oje5q00 [DOI]
- 12. Grossberg A, Mohamed A, Elhalawani H, et al. (2017). Data from Head and Neck Cancer CT Atlas. The Cancer Imaging Archive. 10.7937/k9/tcia.2017.umz8dv6s [DOI]
- 13. Beichel RR, Ulrich EJ, Bauer C, et al. (2015). Data from QIN‐HEADNECK. The cancer imaging archive. 10.7937/k9/tcia.2015.k0f5cgli [DOI]
- 14. Bosch WR, Straube WL, Matthews JW, Purdy JA. (2015). Data From Head‐Neck_Cetuximab. The Cancer Imaging Archive. 10.7937/k9/tcia.2015.7akgjupz [DOI]
- 15. Ger RB, Yang J, Ding Y, et al. (2018). Data from synthetic and phantom MR images for determining deformable image registration accuracy (MRI‐DIR). The Cancer Imaging Archive. 10.7937/k9/tcia.2018.3f08iejt [DOI] [PMC free article] [PubMed]
- 16. Yu CX. Intensity‐modulated arc therapy with dynamic multileaf collimation: an alternative to tomotherapy. Phys Med Biol. 1995;40:1435–1449. [DOI] [PubMed] [Google Scholar]
- 17. Yu CX. Clinical Implementation of intensity‐modulated arc therapy. Int J Radiat Oncol Biol Phys. 2002;53:453–463. [DOI] [PubMed] [Google Scholar]
- 18. Oken MM, Creech RH, Tormey DC, et al. Toxicity and response criteria of the eastern cooperative oncology group. AmJ Clin Oncol. 1982;5:649–655. [PubMed] [Google Scholar]
- 19. AAPM . Task Group No. 66, Quality assurance for computed‐tomography simulators and the computed tomography‐simulation process. American Association of Physicists in Medicine. [ 10.1118/1.1609271] [DOI]
- 20. Bejarano T, De Ornelas CM, Mihaylov IB. (2018). Head‐and‐neck squamous cell carcinoma (HNSCC) patients with CT taken during pre‐treatment, mid‐treatment, and post‐treatment Dataset. The Cancer Imaging Archive. 10.7937/k9/tcia.2018.13upr2xf [DOI]