IKIngerid Skjei Knudtsen

Ingerid Skjei Knudtsen

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Ingerid Skjei Knudtsen has a total of 23 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Reference tissue uptake of [18F]PSMA-1007 in positron emission tomography of recurrent prostate cancer

    by Bendik Skarre Abrahamsen, Ingerid Skjei Knudtsen, Andreas Julius Tulipan, Eivor Hernes, Trond Velde Bogsrud, Kirsten Margrete Selnæs, Håkon Johansen, Tone Frost Bathen, Mattijs Elschot

    Introduction: Semi-quantitative uptake measurements are part of the PROMISE framework for PET/CT interpretation of tracers targeting the prostate-specific membrane antigen (PSMA), where tumor uptake is compared to uptake in reference tissues. [18F]PSMA-1007 has primarily hepatobiliary excretion and it is suggested to use spleen as reference tissue instead of liver. The aim of this study is to investigate the normal range and interpatient variability of [18F]PSMA-1007 uptake in reference tissue for prostate cancer patients, with particular focus on the splenic uptake. Materials and methods: A total of 102 [18F]PSMA-1007 PET/CT scans of prostate cancer patients with biochemical recurrence (BCR) from three hospitals were included in the analysis. SUV max and SUV mean measurements were performed in the blood pool, liver, spleen, and parotid glands. Results are reported as median with percentiles [5th, 95th]. Results: For 15/102 (15%) and 20/102 (20%) of the patients, the SUV max and SUV mean , respectively, were higher in the spleen than in the parotid glands. Median SUV max of the reference tissues were 1.2 [0.6, 2.2] for blood, 15.8 [8.9, 26.2] for spleen, and 26.6 [14.5, 40.4] for the parotid glands. Median SUV mean were 11.9 [6.6, 20.7] for spleen and 19.3 [11.2, 32.1] for the parotid glands. Conclusion: In this multicenter cohort of BCR patients, [18F]PSMA-1007 reference tissue uptake is reported in accordance with the PROMISE protocol. A substantial number of patients present with higher uptake in the spleen than in parotid glands, for reasons that are currently unexplained. Care should be taken when interpreting these values until further studies are conducted. Key Points: Question The normal range for [18F]PSMA-1007 uptake values in PROMISE reference tissues are not well documented, challenging standardized lesion categorization in prostate cancer PET interpretation . Findings In a multicenter cohort, splenic [18F]PSMA-1007 uptake exceeded parotid gland uptake in a notable subgroup (15%), leading to situations where PSMA expression scoring according to the PROMISE framework cannot be applied as intended . Clinical relevance Unexpectedly high splenic uptake relative to parotid gland uptake of [18F]PSMA-1007 in some patients may affect lesion categorization; careful interpretation is advised .

  • Open Access

    Deep learning with uncertainty estimation for automatic tumor segmentation in PET/CT of head and neck cancers: impact of model complexity, image processing and augmentation

    by Bao Ngoc Huynh, Aurora Rosvoll Groendahl, Oliver Tomic, Kristian Hovde Liland, Ingerid Skjei Knudtsen, Frank Hoebers, Wouter J C van Elmpt, Einar Dale, Eirik Malinen, Cecilia Marie Futsaether

    Objective. Target volumes for radiotherapy are usually contoured manually, which can be time-consuming and prone to inter- and intra-observer variability. Automatic contouring by convolutional neural networks (CNN) can be fast and consistent but may produce unrealistic contours or miss relevant structures. We evaluate approaches for increasing the quality and assessing the uncertainty of CNN-generated contours of head and neck cancers with PET/CT as input. Approach. Two patient cohorts with head and neck squamous cell carcinoma and baseline 18F-fluorodeoxyglucose positron emission tomography and computed tomography images (FDG-PET/CT) were collected retrospectively from two centers. The union of manual contours of the gross primary tumor and involved nodes was used to train CNN models for generating automatic contours. The impact of image preprocessing, image augmentation, transfer learning and CNN complexity, architecture, and dimension (2D or 3D) on model performance and generalizability across centers was evaluated. A Monte Carlo dropout technique was used to quantify and visualize the uncertainty of the automatic contours. Main results. CNN models provided contours with good overlap with the manually contoured ground truth (median Dice Similarity Coefficient: 0.75–0.77), consistent with reported inter-observer variations and previous auto-contouring studies. Image augmentation and model dimension, rather than model complexity, architecture, or advanced image preprocessing, had the largest impact on model performance and cross-center generalizability. Transfer learning on a limited number of patients from a separate center increased model generalizability without decreasing model performance on the original training cohort. High model uncertainty was associated with false positive and false negative voxels as well as low Dice coefficients. Significance. High quality automatic contours can be obtained using deep learning architectures that are not overly complex. Uncertainty estimation of the predicted contours shows potential for highlighting regions of the contour requiring manual revision or flagging segmentations requiring manual inspection and intervention.


  • A comparison of methods for fully automatic segmentation of tumors and involved nodes in PET/CT of head and neck cancers

    by Aurora Rosvoll Groendahl, Ingerid Skjei Knudtsen, Bao Ngoc Huynh, Martine Mulstad, Yngve Mardal Mardal Moe, Franziska Knuth, Oliver Tomic, Ulf Geir Indahl, Turid Torheim, Einar Dale, et al.

    Target volume delineation is a vital but time-consuming and challenging part of radiotherapy, where the goal is to deliver sufficient dose to the target while reducing risks of side effects. For head and neck cancer (HNC) this is complicated by the complex anatomy of the head and neck region and the proximity of target volumes to organs at risk. The purpose of this study was to compare and evaluate conventional PET thresholding methods, six classical machine learning algorithms and a 2D U-Net convolutional neural network (CNN) for automatic gross tumor volume (GTV) segmentation of HNC in PET/CT images. For the latter two approaches the impact of single versus multimodality input on segmentation quality was also assessed. 197 patients were included in the study. The cohort was split into training and test sets (157 and 40 patients, respectively). Five-fold cross-validation was used on the training set for model comparison and selection. Manual GTV delineations represented the ground truth. Tresholding, classical machine learning and CNN segmentation models were ranked separately according to the cross-validation Sørensen–Dice similarity coefficient (Dice). PET thresholding gave a maximum mean Dice of 0.62, whereas classical machine learning resulted in maximum mean Dice scores of 0.24 (CT) and 0.66 (PET; PET/CT). CNN models obtained maximum mean Dice scores of 0.66 (CT), 0.68 (PET) and 0.74 (PET/CT). The difference in cross-validation Dice between multimodality PET/CT and single modality CNN models was significant (p ≤ 0.0001). The top-ranked PET/CT-based CNN model outperformed the best-performing thresholding and classical machine learning models, giving significantly better segmentations in terms of cross-validation and test set Dice, true positive rate, positive predictive value and surface distance-based metrics (p ≤ 0.0001). Thus, deep learning based on multimodality PET/CT input resulted in superior target coverage and less inclusion of surrounding normal tissue.