Takeshi Kondo
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
Research Subjects
Takeshi Kondo has a total of 184 co-authors. Here are the top co-authors by number of shared publications:
- Sohei Kitazawa
Department of Molecular Pathology Ehime University Graduate School of Medicine, Shitsukawa Toon City Japan
13 shared publications Yasuhiro Ueno
Division of Legal Medicine, Department of Community Medicine and Social Healthcare Science, Kobe University Graduate School of Medicine, Kobe, Japan
10 shared publicationsMigiwa Asano
Department of Legal Medicine, Ehime University Graduate School of Medicine, Toon, Japan
9 shared publicationsMotonori Takahashi
Division of Legal Medicine, Department of Community Medicine and Social Healthcare Science, Kobe University Graduate School of Medicine, Kobe, Japan
9 shared publicationsAzumi Kuse
Division of Legal Medicine, Department of Community Medicine and Social Healthcare Science, Kobe University Graduate School of Medicine, Kobe, Japan
8 shared publicationsKanako Nakagawa
Division of Legal Medicine, Department of Community Medicine and Social Healthcare Science, Kobe University Graduate School of Medicine, Kobe, Japan
8 shared publications
Latest Publications
- 11 February 2026
- Published by Elsevier in Journal of Forensic and Legal Medicine
- Vol. 118, 103090
Family satisfaction with forensic autopsy in Japan: The role of medical explanation and supportive environment
- 11 February 2026
- Published by Elsevier in Journal of Forensic and Legal Medicine
- Vol. 118, 103090
Family satisfaction with forensic autopsy in Japan: The role of medical explanation and supportive environment
- Open Access
- 28 September 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (19) , 10513
Deep Learning-Based Detection of Intracranial Hemorrhages in Postmortem Computed Tomography: Comparative Study of 15 Transfer-Learned Models
With the increasing use of postmortem imaging, deep learning (DL)-based automated analysis may assist in the detection of intracranial hemorrhages. However, limited postmortem data complicate model training. This study aims to assess the accuracy of DL models in detecting intracranial hemorrhages in postmortem head computed tomography (CT) scans using transfer learning. A total of 75,000 labeled head CT images from the Radiological Society of North America Intracranial Hemorrhage Detection Challenge serve as the training data for the 15 DL models. Each model is fine-tuned via transfer learning. A total of 134 postmortem cases with hemorrhage status confirmed by autopsy serve as the external test set. Model performance is evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, training time, inference time, and number of parameters. Spearman’s rank correlation coefficients are calculated for these metrics. DenseNet201 achieves the highest AUC (0.907), with the AUCs of the 15 models ranging from 0.862 to 0.907. A longer inference time moderately correlates with higher AUC (Spearman’s ρ = 0.586, p = 0.022), whereas the number of parameters is not positively correlated with performance (ρ = −0.472, p = 0.076). The sensitivity and specificity are 0.828 and 0.871, respectively. Transfer learning using a large non-postmortem dataset enables accurate intracranial hemorrhage detection using postmortem CT, potentially reducing the autopsy workload. The results demonstrate that models with fewer parameters often perform comparably to more complex models, emphasizing the need to balance accuracy with computational efficiency.
- 28 September 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (19) , 10513
Open AccessDeep Learning-Based Detection of Intracranial Hemorrhages in Postmortem Computed Tomography: Comparative Study of 15 Transfer-Learned Models
With the increasing use of postmortem imaging, deep learning (DL)-based automated analysis may assist in the detection of intracranial hemorrhages. However, limited postmortem data complicate model training. This study aims to assess the accuracy of DL models in detecting intracranial hemorrhages in postmortem head computed tomography (CT) scans using transfer learning. A total of 75,000 labeled head CT images from the Radiological Society of North America Intracranial Hemorrhage Detection Challenge serve as the training data for the 15 DL models. Each model is fine-tuned via transfer learning. A total of 134 postmortem cases with hemorrhage status confirmed by autopsy serve as the external test set. Model performance is evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, training time, inference time, and number of parameters. Spearman’s rank correlation coefficients are calculated for these metrics. DenseNet201 achieves the highest AUC (0.907), with the AUCs of the 15 models ranging from 0.862 to 0.907. A longer inference time moderately correlates with higher AUC (Spearman’s ρ = 0.586, p = 0.022), whereas the number of parameters is not positively correlated with performance (ρ = −0.472, p = 0.076). The sensitivity and specificity are 0.828 and 0.871, respectively. Transfer learning using a large non-postmortem dataset enables accurate intracranial hemorrhage detection using postmortem CT, potentially reducing the autopsy workload. The results demonstrate that models with fewer parameters often perform comparably to more complex models, emphasizing the need to balance accuracy with computational efficiency.
- Open Access
- 5 September 2025
- Published by MDPI AG in Preprints.org
Deep Learning–Based Detection of Intracranial Hemorrhages in Postmortem Computed Tomography: Comparative Study of 15 Transfer-Learned Models
With the increasing use of postmortem imaging, deep learning (DL)-based automated analysis may assist in detecting intracranial hemorrhages. However, limited postmortem data complicate model training. This study aims to assess the accuracy of DL models in detecting intracranial hemorrhages in postmortem head computed tomography (CT) scans using transfer learning. 75,000 labeled head CT images from the Radiological Society of North America Intracranial Hemorrhage Detection Challenge serve as training data for 15 DL models. Each model is fine-tuned via transfer learning. A total of 124 postmortem cases with hemorrhage status confirmed by autopsy serve as the external test set. The model performance is evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, training time, inference time, and number of parameters. Spearman’s rank correlation coefficients are calculated for these metrics. DenseNet201 achieves the highest AUC (0.907), with the AUCs of the 15 models ranging from 0.862 to 0.907. Longer inference time moderately correlates with higher AUC (Spearman’s ρ = 0.586, p = 0.022), whereas the number of parameters is not positively correlated with performance (ρ = -0.472, p = 0.076). The sensitivity and specificity are 0.906 and 0.986, respectively. Transfer learning using a large non-postmortem dataset enables accurate intracranial hemorrhage detection on postmortem CT, potentially reducing autopsy workload. The results demonstrate that models with fewer parameters often perform comparably to more complex models, emphasizing the need to balance accuracy with computational efficiency.
- 5 September 2025
- Published by MDPI AG in Preprints.org
Open AccessDeep Learning–Based Detection of Intracranial Hemorrhages in Postmortem Computed Tomography: Comparative Study of 15 Transfer-Learned Models
With the increasing use of postmortem imaging, deep learning (DL)-based automated analysis may assist in detecting intracranial hemorrhages. However, limited postmortem data complicate model training. This study aims to assess the accuracy of DL models in detecting intracranial hemorrhages in postmortem head computed tomography (CT) scans using transfer learning. 75,000 labeled head CT images from the Radiological Society of North America Intracranial Hemorrhage Detection Challenge serve as training data for 15 DL models. Each model is fine-tuned via transfer learning. A total of 124 postmortem cases with hemorrhage status confirmed by autopsy serve as the external test set. The model performance is evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, training time, inference time, and number of parameters. Spearman’s rank correlation coefficients are calculated for these metrics. DenseNet201 achieves the highest AUC (0.907), with the AUCs of the 15 models ranging from 0.862 to 0.907. Longer inference time moderately correlates with higher AUC (Spearman’s ρ = 0.586, p = 0.022), whereas the number of parameters is not positively correlated with performance (ρ = -0.472, p = 0.076). The sensitivity and specificity are 0.906 and 0.986, respectively. Transfer learning using a large non-postmortem dataset enables accurate intracranial hemorrhage detection on postmortem CT, potentially reducing autopsy workload. The results demonstrate that models with fewer parameters often perform comparably to more complex models, emphasizing the need to balance accuracy with computational efficiency.
- 28 July 2022
- Published by Elsevier in Legal Medicine
- Vol. 59, 102127
Immunohistochemical analysis of CD31 expression in myocardial tissues from autopsies of patients with ischemic heart disease
- 28 July 2022
- Published by Elsevier in Legal Medicine
- Vol. 59, 102127
Immunohistochemical analysis of CD31 expression in myocardial tissues from autopsies of patients with ischemic heart disease
- 1 February 2022
- Published by Elsevier in Legal Medicine
- Vol. 54, 102003
Immunohistochemical analysis of vimentin expression in myocardial tissue from autopsy cases of ischemic heart disease
- 1 February 2022
- Published by Elsevier in Legal Medicine
- Vol. 54, 102003
Immunohistochemical analysis of vimentin expression in myocardial tissue from autopsy cases of ischemic heart disease
- 1 February 2022
- Published by Elsevier in Legal Medicine
- Vol. 54, 101997
Immunohistochemical analysis of von Willebrand factor expression in myocardial tissues from autopsies of patients with ischemic heart disease
- 1 February 2022
- Published by Elsevier in Legal Medicine
- Vol. 54, 101997
Immunohistochemical analysis of von Willebrand factor expression in myocardial tissues from autopsies of patients with ischemic heart disease
- Open Access
- 1 January 2022
- Published by University of Tokushima Faculty of Medicine in The Journal of Medical Investigation
- Vol. 69 (3.4) , 287-293
An attempt to create a treatment algorithm of central adrenal insufficiency using CRH test, DHEA-S and clinical evaluation
Objective : To examine diagnostic performance of corticotropin-releasing hormone (CRH) test combined with baseline dehydroepiandrosterone sulfate (DHEA-S) in patients with a suspect of central adrenal insufficiency. Methods : Patients (n=215) requiring daily or intermittent hydrocortisone replacement, or no replacement were retrospectively checked with their peak cortisol after CRH test and baseline DHEA-S. Results : None of 106 patients with the peak cortisol ≥ 17.5 µg / dL after CRH test required replacement, and all 64 patients with the peak cortisol < 10.0 µg / dL required daily replacement. Among 8 patients with 10.0 µg / dL ≤ the peak cortisol < 17.5 µg / dL and baseline DHEA-S below the reference range, 6 patients required daily replacement and 1 patient was under intermittent replacement. Among 37 patients with 10.0 µg / dL ≤ the peak cortisol < 17.5 µg / dL and baseline DHEA-S within the reference range, 10 and 6 patients were under intermittent and daily replacement, respectively. Conclusions : No patients with the peak cortisol ≥ 17.5 µg / dL required hydrocortisone replacement, and all patients with the peak cortisol below 10.0 µg / dL required daily replacement. Careful clinical evaluation was required to determine requirement for replacement in patients with 10.0 µg / dL ≤ the peak cortisol < 17.5 µg / dL even in combination with baseline DHEA-S. J. Med. Invest. 69 : 287-293, August, 2022
- 1 January 2022
- Published by University of Tokushima Faculty of Medicine in The Journal of Medical Investigation
- Vol. 69 (3.4) , 287-293
Open AccessAn attempt to create a treatment algorithm of central adrenal insufficiency using CRH test, DHEA-S and clinical evaluation
Objective : To examine diagnostic performance of corticotropin-releasing hormone (CRH) test combined with baseline dehydroepiandrosterone sulfate (DHEA-S) in patients with a suspect of central adrenal insufficiency. Methods : Patients (n=215) requiring daily or intermittent hydrocortisone replacement, or no replacement were retrospectively checked with their peak cortisol after CRH test and baseline DHEA-S. Results : None of 106 patients with the peak cortisol ≥ 17.5 µg / dL after CRH test required replacement, and all 64 patients with the peak cortisol < 10.0 µg / dL required daily replacement. Among 8 patients with 10.0 µg / dL ≤ the peak cortisol < 17.5 µg / dL and baseline DHEA-S below the reference range, 6 patients required daily replacement and 1 patient was under intermittent replacement. Among 37 patients with 10.0 µg / dL ≤ the peak cortisol < 17.5 µg / dL and baseline DHEA-S within the reference range, 10 and 6 patients were under intermittent and daily replacement, respectively. Conclusions : No patients with the peak cortisol ≥ 17.5 µg / dL required hydrocortisone replacement, and all patients with the peak cortisol below 10.0 µg / dL required daily replacement. Careful clinical evaluation was required to determine requirement for replacement in patients with 10.0 µg / dL ≤ the peak cortisol < 17.5 µg / dL even in combination with baseline DHEA-S. J. Med. Invest. 69 : 287-293, August, 2022
- 1 July 2021
- Published by Elsevier in Legal Medicine
- Vol. 51, 101897
Immunohistochemical analysis of thrombomodulin expression in myocardial tissue from autopsy cases of ischemic heart disease
- 1 July 2021
- Published by Elsevier in Legal Medicine
- Vol. 51, 101897
Immunohistochemical analysis of thrombomodulin expression in myocardial tissue from autopsy cases of ischemic heart disease
- Open Access
- 20 April 2021
- Published by Springer Nature in Surgical Case Reports
- Vol. 7 (1) , 1-6
A case of contralateral inguinal lymph node metastases from breast cancer
Background: Breast cancer is well known to tends to invade through the lymphatic chains mainly to the axillary and subclavian nodes or occasionally to the internal mammary nodes. However, inguinal lymph node metastasis from breast cancer is extremely rare. Case presentation: We have experienced a case of an 82-year-old woman showing left inguinal lymph node metastases from right breast cancer. Previously, she had received five times abdominal operations and left artificial bone head replacement for metamorphous hip-joint disease. Although the metastases were firstly detected 46 months after the breast surgery, they had already existed at the time of the breast operation, which was retrospectively re-evaluated by CT examination. The progression pattern of inguinal lymph node metastases had much correlated with that of the breast cancer. She underwent inguinal lymph node dissections. Pathological findings revealed them being compatible with breast cancer origin. Conclusions: This is the sixth case having been reported in English literature. Besides, this is the first case showing the contralateral spread to the primary breast cancer. One of the causes of this complex metastatic pattern is thought be ascribed to the previously performed prolific abdominal operations.
- 20 April 2021
- Published by Springer Nature in Surgical Case Reports
- Vol. 7 (1) , 1-6
Open AccessA case of contralateral inguinal lymph node metastases from breast cancer
Background: Breast cancer is well known to tends to invade through the lymphatic chains mainly to the axillary and subclavian nodes or occasionally to the internal mammary nodes. However, inguinal lymph node metastasis from breast cancer is extremely rare. Case presentation: We have experienced a case of an 82-year-old woman showing left inguinal lymph node metastases from right breast cancer. Previously, she had received five times abdominal operations and left artificial bone head replacement for metamorphous hip-joint disease. Although the metastases were firstly detected 46 months after the breast surgery, they had already existed at the time of the breast operation, which was retrospectively re-evaluated by CT examination. The progression pattern of inguinal lymph node metastases had much correlated with that of the breast cancer. She underwent inguinal lymph node dissections. Pathological findings revealed them being compatible with breast cancer origin. Conclusions: This is the sixth case having been reported in English literature. Besides, this is the first case showing the contralateral spread to the primary breast cancer. One of the causes of this complex metastatic pattern is thought be ascribed to the previously performed prolific abdominal operations.
- 1 February 2021
- Published by Elsevier in Legal Medicine
- Vol. 48, 101812
Autopsy case of Rosai–Dorfman disease presenting as fibrinous pericarditis
- 1 February 2021
- Published by Elsevier in Legal Medicine
- Vol. 48, 101812
Autopsy case of Rosai–Dorfman disease presenting as fibrinous pericarditis