Machine Learning Enhanced Virtual Autopsy

O’Sullivan, Shane; Holzinger, Andreas; Zatloukalc, Kurt; Saldiva, Paulo Hilário Nascimento; Sajid, Mohammed Imran; Wichmann, Dominic

doi:10.4322/acr.2017.037

Cited by 22 publications

(20 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The big advantage of such systems would include not only explainability, but deeper understanding and replicability [8]. Most of all, this would increase acceptance and trust, which is mandatory in safety-critical systems [12], and desirable in many applications (e.g., in medical robotics [19], Ambient Assisted Living [23], Enterprise decision making [4], etc.). First steps have been taken towards making these systems understandable to their users, by providing textual and visual explanations [13], [22] (see Figures 2 and 3).…”

Section: Laysan Albatrossmentioning

confidence: 99%

Explainable AI: The New 42?

Goebel

Chander

Holzinger

et al. 2018

Lecture Notes in Computer Science

Self Cite

224

View full text Add to dashboard Cite

their un-debuggability, and their inability to "explain" their decisions in a human understandable and reconstructable way. So while AlphaGo or DeepStack can crush the best humans at Go or Poker, neither program has any internal model of its task; its representations defy interpretation by humans, there is no mechanism to explain their actions and behaviour, and furthermore, there is no obvious instructional value. .. the high performance systems can not help humans improve. Even when we understand the underlying mathematical scaffolding of current machine learning architectures, it is often impossible to get insight into the internal working of the models; we need explicit modeling and reasoning tools to explain how and why a result was achieved. We also know that a significant challenge for future AI is contextual adaptation, i.e., systems that incrementally help to construct explanatory models for solving real-world problems. Here it would be beneficial not to exclude human expertise, but to augment human intelligence with artificial intelligence.

show abstract

Section: Laysan Albatrossmentioning

confidence: 99%

Explainable AI: The New 42?

Goebel

Chander

Holzinger

et al. 2018

Lecture Notes in Computer Science

Self Cite

224

View full text Add to dashboard Cite

show abstract

“…This is important because this actually causes limitations in machine-based identification of brain regions and pathologies. Consequently, providing high-resolution images in general, and MRI data in particular, not only supports the development of ML algorithms for image analysis per se, but enables us to reach a further step towards diagnosis assistance [77,78].…”

Section: The Role Of Computer Science Ai and ML In Integrating Virtumentioning

confidence: 93%

The role of artificial intelligence and machine learning in harmonization of high-resolution post-mortem MRI (virtopsy) with respect to brain microstructure

et al. 2019

Self Cite

View full text Add to dashboard Cite

Enhanced resolution of 7 T magnetic resonance imaging (MRI) scanners has considerably advanced our knowledge of structure and function in human and animal brains. Post-industrialized countries are particularly prone to an everincreasing number of ageing individuals and ageing-associated neurodegenerative diseases. Neurodegenerative diseases are associated with volume loss in the affected brain. MRI diagnoses and monitoring of subtle volume changes in the ageing/diseased brains have the potential to become standard diagnostic tools. Even with the superior resolution of 7 T MRI scanners, the microstructural changes comprising cell types, cell numbers, and cellular processes, are still undetectable. Knowledge of origin, nature, and progression for microstructural changes are necessary to understand pathogenetic stages in the relentless neurodegenerative diseases, as well as to develop therapeutic tools that delay or stop neurodegenerative processes at their earliest stage. We illustrate the gap in resolution by comparing the identical regions of the post-mortem in situ 7 T MR images (virtual autopsy or virtopsy) with the histological observations in serial sections through the same brain. We also described the protocols and limitations associated with these comparisons, as well as the necessity of supercomputers and data management for "Big data". Analysis of neuron and/ or glial number by using a body of mathematical tools and guidelines (stereology) is time-consuming, cumbersome, and still restricted to trained human investigators. Development of tools based on machine learning (ML) and artificial intelligence (AI) could considerably accelerate studies on localization, onset, and progression of neuron loss. Finally, these observations could disentangle the mechanisms of volume loss into stages of reversible atrophy and/or irreversible fatal cell death. This AI-and ML-based cooperation between virtopsy and histology could bridge the present gap between virtual reality and neuropathology. It could also culminate in the creation of an imaging-associated comprehensive database. This database would include genetic, clinical, epidemiological, and technical aspects that could help to alleviate or even stop the adverse effects of neurodegenerative diseases on affected individuals, their families, and society.

show abstract

“…At night, ultrasound can also be used in the darkness of a medical tent when lighting is not available. In brief, such use of portable ultrasound-guided biopsy might have been very useful in these outbreaks (O'Sullivan et al 2017). In addition, ultrasound-guided needle biopsy could have been highly effective in quickly identifying the cause of the outbreak (O'Sullivan et al 2017).…”

Section: Virtual Autopsy (Virtopsy)mentioning

confidence: 99%

“…This approach may have reduced the number of Ebola cases worldwide. Moreover, this could have potentially enabled an earlier diagnosis linking Zika virus to microcephaly recently in Northeast Brazil (O'Sullivan et al 2017). If the diagnosis was made earlier, additional cases may have been prevented.…”

Section: Virtual Autopsy (Virtopsy)mentioning

confidence: 99%

Virtual autopsy and community engagement for outbreak response in Africa: traditional, religious and sociocultural perspectives

O’Sullivan

Sajid

Agusto

et al. 2018

Egypt J Forensic Sci

Self Cite

View full text Add to dashboard Cite

Background: In response to outbreaks, the manner in which the medical field engages with the process of death is significantly important, not only in the pathological but also in the sociocultural aspect. Certain response-methodology used could be a critical catalyst for community resistance to post-mortem examinations (e.g. hiding corpses, violence), considering how the deceased are handled in different cultures. We reviewed the traditional, religious and sociocultural aspects of post-mortem response-methodology in Africa. With no means to take blood samples, innovative sampling techniques were developed and frequently adopted in post-mortem cases. However, this approach can be met with resistance. Main body: To avoid resistance, virtual autopsy (virtopsy) involving a non-invasive or minimally invasive imaging-guided biopsy (e.g. portable ultrasound guidance) is recommended as a proper entry point to negotiations around examination of the deceased. We outline how adapting virtopsy and community engagement strategies can have many benefits to communicable disease prevention or control, public health and communities in Africa and elsewhere. This paper provides a completely new perspective by applying imaging autopsy to field studies. These are usually hospital-based procedures. Many countries can train their response and military personnel on how to perform a safe straightforward virtopsy procedure in an outbreak situation. Just as relevant, Islam is now the fastest growing religion in the world-virtopsy might appease fears and increase the autopsy rate among large Muslim populations (in Africa and elsewhere).Conclusions: Virtopsy use and its application might have proved very useful in the recent outbreak of Ebola in 2014 and microcephaly associated with Zika virus. Use of virtopsy may have reduced the number of Ebola cases worldwide and enabled an earlier diagnosis linking Zika virus to microcephaly during the recent outbreak in Northeast Brazil. If the diagnosis was made earlier, additional cases of Ebola or Zika may have been prevented. The issue of invasive autopsy was in all likelihood one of the greatest factors that drove Ebola underground; it prevented communities from cooperating with non-native response teams and contributed to the excessive duration and length of the outbreak. Healers may unknowingly infect others (via scarification) and themselves (via contact with others). Adapting sociocultural engagement methods and conducting a virtopsy may be more acceptable, thereby ameliorating the spread of a deadly virus more efficiently.

show abstract

Machine Learning Enhanced Virtual Autopsy

Cited by 22 publications

References 6 publications

Explainable AI: The New 42?

Explainable AI: The New 42?

The role of artificial intelligence and machine learning in harmonization of high-resolution post-mortem MRI (virtopsy) with respect to brain microstructure

Virtual autopsy and community engagement for outbreak response in Africa: traditional, religious and sociocultural perspectives

Contact Info

Product

Resources

About